JP6383789B2 - Large-scale biomolecule analysis using sequence tags - Google Patents

Description

(Cross-reference of related applications)
This application claims priority to US Provisional Application No. 61 / 841,878, filed July 1, 2013, and 62 / 001,580, filed May 21, 2014. These provisional applications are hereby incorporated by reference in their entirety.

(Sequence Listing)
This application includes an ASCII format sequence listing submitted by EFS-Web, the entire contents of which are incorporated herein by reference. The copy of ASCII was created on June 27, 2014, and the file name is 848US00-SL-ST25. txt, and the file size is 2 kilobytes. No new content has been added.

  Large scale DNA sequencing is rapidly evolving in diagnostic and predictive applications as speed and convenience increase and cost per base is reduced [eg Ding et al, Nature, 481 (7342): 506-510 (2012); Chiu et al, Brit. Med. J. et al. 342: c7401 (2011); and Ku et al, Anals of Neurology, 71 (1): 5-14 (2012), etc.]. In particular, the profile of nucleic acids encoding immune molecules such as T cell or B cell receptors or their constituent molecules has a wealth of information about the health or disease state of the organism, so this profile Have been proposed to be used as indicators for the diagnosis or prediction of various conditions [eg, Faham and Willis, US Pat. Nos. 8,236,503 and 8,628927; Freeman et al, Genome Research, 19: 1817-1824 (2009); Han et al, J. MoI. Immunol. , 182 (1001): 42.6 (2009); Boyd et al, Sci. Transl. Med. , 1 (12): 12ra23 (2009); He et al, Oncotarget (March 8, 2011)].

  For example, patients treated for many cancers often have minimal residual disease (MRD) associated with cancer. That is, on a clinical scale, for some reason, a small population of cancer cells can escape escape and remain, even if the patient's disease is said to be completely ameliorated by treatment. The type and size of this remaining population is an important prognostic factor related to continued therapy in patients [see, for example, Campana, Hematol. Oncol. Clin. North Am. , 23 (5): 1083-1098 (2009); Buccino et al, Blood, 119 (2): 332-341 (2012)]. As a result, a number of techniques for evaluating this population have been developed, such as techniques based on flow cytometry, in situ hybridization, cytogenetics, and amplification of nucleic acid markers [eg, Buccino et al, Current Opinion in Oncology, 21 : 582-588 (2009); and van Dongen et al, Leukemia, 17 (12): 2257-2317 (2003), etc.]. Amplification of reconstituted nucleic acids encoding immunoreceptor fragments (ie, clonotypes) of T cells and / or B cells is a molecule in which such clonotypes are typically associated with cancer cells. Having a unique sequence that can be provided as a tag is particularly useful for the assessment of MRD in leukemias and lymphomas. Due in part to the fact that amplification is highly multiplexed and difficult to develop, measurements are usually made by amplification and sequencing of the nucleic acid encoding the short chain of the receptor. As the scale of multiplexing increases, we face several problems, such as increasing the probability of false positive amplification by mishybridization, formation of primer dimers, and biased display of sequences due to different amplification rates. [For example, Elnifro et al, Clinical Microbiology Reviews, 13 (4): 559-570 (2000)]. Furthermore, whether for sequence analysis, sample tracking, or detection of contaminants, the similarity of the target sequence and the incorporation of the sequence tag into the amplified sequence adversely affects the challenges associated with large-scale amplification described above. It is exerting. These challenges are due to the development of large-scale single-reaction amplification of multiple immunoreceptor chains that is very effective in reducing the number of separate assays required to evaluate nucleic acid sequences that correlate with microlesions. It has been resolved.

  From these perspectives, more efficient methods are available that evaluate selected nucleic acids, such as exons or clonotypes of oncogenes that encode sets of immunoreceptor chains, in a single reaction. This is very advantageous.

  The present invention relates to large scale amplification methods in a single reaction, in particular large scale DNA sequences after polymerase chain reaction (PCR) of a target polynucleotide population such as a reconstituted nucleic acid encoding an immunoreceptor chain. The present invention relates to a method for identifying them using a determination method. The present invention includes monitoring of minimal residual lesions in cancer by applying the method described above. The present invention is exemplified in numerous implementations and uses, some of which are summarized throughout the specification.

  In some embodiments, the present invention is directed to a method of generating a nucleic acid profile encoding a group of biomolecules of interest, such as an immunoreceptor molecule. In one aspect, the method of the invention comprises adding a sequence tag to a selected group of nucleic acids in a sample to produce a tag-nucleic acid complex, amplifying the tag-nucleic acid complex, and amplifying. Sequencing the tag-nucleic acid complex to provide a sequence read that each includes both the tag sequence and the nucleic acid sequence to generate a nucleic acid profile. In some embodiments, the addition of a sequence tag can be accomplished by one or more successive steps related to primer extension and primer removal, after which the resulting product is converted into the forward direction. Further amplification can be performed without bias by using reverse primers.

  In some embodiments, the present invention is directed to a method for detecting and measuring sample contamination, eg, carry-over contamination, with material from different samples. In one embodiment, such methods for detecting contamination in individuals monitored for minimal residual disease include: (a) obtaining a tissue sample from the individual; (b) an oncogene molecule or reconstituted Adding a sequence tag to the nucleic acid to produce a tag-nucleic acid complex (at least one nucleic acid or a copy thereof has a different sequence tag added, and an oncogene molecule is characterized by an individual's cancer (C) amplifying the tag-nucleic acid complex; (d) sequencing the tag-nucleic acid complex sample and having an error rate, the tag sequence and the oncogene sequence or Providing a sequence read comprising the reconstituted nucleic acid sequence; (e) comparing the tag sequence to a tag sequence determined separately from another tissue sample; (f) one or more tag sequences Presence of identity Absence, and / or by the degree of contamination includes evaluating the tag sequence which is separately determined from the other tissue samples, and.

  In another aspect, the present invention is directed to a method as described above for generating a clonotype profile based on at least two B cell receptor chains, the method comprising two or more B cell receptor chains. Amplifying the encoded target nucleic acid in a single reaction. In another aspect, such methods are used to monitor minimal residual disease in B cell cancer.

  In another aspect, the present invention is directed to a method as described above for generating a clonotype profile based on at least two T cell receptor chains, the method comprising two or more T cell receptor chains. Amplifying the encoded target nucleic acid in a single reaction. In another aspect, such methods are used to monitor minimal residual disease in T cell cancer.

  These above-characterized aspects of the invention, as well as other aspects, are illustrated by numerous examples and applications. Some of them are illustrated in the drawings and characterized in the following claims. However, each of the above summary is not intended to describe each embodiment or every implementation described in the present invention.

The novel features of the invention are set forth with particularity in the appended claims. A better understanding of the features and advantages of the present invention will be obtained by reference to the following detailed description that sets forth illustrative embodiments, in which the principles of the invention are utilized, and the accompanying drawings of which:
Various embodiments of the invention are shown in schematic form. Various embodiments of the invention are shown in schematic form. Various embodiments of the invention are shown in schematic form. A method of generating a reconstituted nucleic acid template having a predetermined length (when using or not using a sequence tag) is shown. Various methods are shown for adding unique sequence tags to virtually all target sequences in a sample. Various methods are shown for adding unique sequence tags to virtually all target sequences in a sample. Various methods are shown for adding unique sequence tags to virtually all target sequences in a sample. Various methods are shown for adding unique sequence tags to virtually all target sequences in a sample. Various methods are shown for adding unique sequence tags to virtually all target sequences in a sample. Various methods are shown for adding unique sequence tags to virtually all target sequences in a sample. Various methods are shown for adding unique sequence tags to virtually all target sequences in a sample. FIG. 2 shows a schematic diagram of an embodiment of the present invention for generating a clonotype profile from a nucleic acid sequence encoding an IgH chain. FIG. 2 shows a schematic diagram of an embodiment of the present invention for generating a clonotype profile from a nucleic acid sequence encoding an IgH chain. FIG. 5 illustrates the use of sequence tags to determine clonotype sequences from sequence reads. FIG. 4 illustrates the use of sequence tags in embodiments where multiple different sequence tags are added to the same target polynucleotide or a copy thereof. The concept of chronotypes in the arrangement space and distance between closely related chronotypes is shown. FIG. 5 is a flow chart illustrating one embodiment of a method for distinguishing truly different chronotypes from chronotypes that differ (to be combined) simply due to alignment errors. Fig. 4 shows a number of function forms used in one embodiment for determining whether or not to associate related chronotypes. Fig. 4 illustrates the use of a sequence tree in a method for combining sequence reads. Fig. 4 illustrates the use of a sequence tree in a method for combining sequence reads.

  Unless otherwise stated, the practice of the present invention uses conventional methods and descriptions of molecular biology (such as recombinant methods), bioinformatics, cell biology, and biochemistry that are within the skill of the art. can do. Such conventional methods include, but are not limited to, blood cell sample collection and analysis, and nucleic acid sequencing and analysis. Specific examples of suitable techniques can be obtained by reference to the following examples. However, other equivalent conventional methods can of course be used. Such conventional methods and descriptions are described in standard laboratory manuals such as Genome Analysis: A Laboratory Manual Series (Vols. I-IV); PCR Primer: A Laboratory Manual; , Published by Cold Spring Harbor Laboratory Press).

  In one aspect, the present invention provides such immunoreceptors by performing high-throughput sequencing on amplification products or amplicons after large-scale multiplex amplification of nucleic acids encoding multiple immunoreceptor chains. It is aimed at a method for generating a chronotype profile of a chain. In some embodiments, the present invention includes any of a series of extending primers, removing unextended or unincorporated primers, and amplification (eg, by PCR) or further primer extension. By incorporating a new primer for this purpose, the disadvantages common to multiplex amplification are overcome. Such a process also makes it possible to use sequence tags. Not using sequence tags leads to non-specific or false positive amplification. In another aspect, the sequence tag is utilized in embodiments related to clinical applications, particularly for analysis of minimal residual disease (MRD), for example, MRD analysis of patient samples undergoing treatment for cancer. . The sequence tag incorporated into the sequence read provides an effective means for chronotype determination, while at the same time different samples from the same patient sample brought in from previous assays or examined in the same laboratory. By detecting the presence or absence of a sequence tag derived from a sample, a convenient means for detecting carryover contamination is provided. Chronotypes based on the sequence of reconstituted nucleic acids encoding multiple B cell receptor (BCR) chains by using high-throughput next generation sequencing methods after a single amplification reaction Of particular interest are methods of generating profiles. Similarly, based on reconstituted nucleic acid sequences encoding multiple T cell receptor (TCR) chains, using a single amplification reaction followed by high-throughput next generation sequencing. Of particular interest is a method of generating a chronotype profile. The methods of the present invention can also be applied to other large scale amplification and sequencing methods of interest, for example other sets of nucleic acids, such as exon sets of oncogenes. In these embodiments, the sequence tag allows both monitoring of carryover contamination and more sensitive determination of the nucleotide sequence of the target polynucleotide in an error prone sequencing method. . Similarly, in these aspects, the set of sequence tags (discussed more fully below) is typically much larger than the number of target polynucleotides in the sample and added to the target polynucleotide. Since the difference in sequence between the sequence tags is sufficiently large, in practice, the sequence of one tag is not converted to another tag due to sequencing errors.

  One embodiment of the present invention is illustrated in FIG. 1A. In the reaction mixture, a first set of primers (22) [a first set of each primer comprising a receptor specific portion (16) and a 5′-non-complementary portion (15 ) And annealed to one end of the target polynucleotide (10) [after thawing of the target polynucleotide (10)], a second set of primers (24) [each primer of the second set is receptor specific And a 5′-non-complementary portion comprising a sequence tag (14) and a second primer binding site (12)] annealing to another end of the target polynucleotide (10) To do. In some embodiments, as described below, the non-complementary portion (15) of primer (22) may also include a sequence tag. In some situations, two types of short sequence tags may be more convenient than one longer sequence tag with similar diversity. Thus, for example, two types of 8-mer random nucleotide sequence tags are less likely to cause false positives and primer dimer formation than one type of 16-mer random nucleotide sequence tags. The target polynucleotide (10) is typically encoded in the chain or chain portion of a T cell receptor (TCR) or B cell receptor (eg, a portion of an IgH chain or IgK chain) Alternatively, it is a somatically reconstituted nucleic acid derived from a B cell. Thus, in some embodiments, the receptor specific portions of primers (22) and (24) may be specific for the V region sequence and the J region sequence, respectively, or in other embodiments The reverse is also possible.

  In some embodiments, the target polynucleotide (10) is a rearranged nucleic acid encoding a portion of an immunoreceptor molecule, 16S of the microbial population, including but not limited to a sequence profile desired. rDNA, metagenomic amplification products of genes encoding industrially or pharmaceutically important proteins (such as enzymes), human or animal genes, and / or exons related to specific diseases such as cancer or infectious diseases Or a complex mixture of nucleic acids. In embodiments involving reconstituted nucleic acid encoding an immunoreceptor, usually at least a portion of the V, D, or J region is between the two binding positions of the first and second primer sets. Exists. In some embodiments, between the two binding positions of the first and second primer sets, at least an IgH VDJ rearrangement sequence, an IgH DJ rearrangement sequence, an IgK VJ rearrangement sequence, an IgL VJ rearrangement, TCR β VDJ rearrangement, TCR β DJ rearrangement, TCR α VJ rearrangement, TCR γ VJ rearrangement, TCR δ VDJ rearrangement, or TCR δ VD There is a part of the rearranged sequence. In some embodiments, at least between the two binding positions of the first and second primer sets is at least an IgH VDJ rearrangement sequence, an IgH DJ rearrangement sequence, an IgK VJ rearrangement sequence, or an IgL There are parts of the VJ rearrangement sequence. In some embodiments, at least between the two binding positions of the first and second primer sets, at least a TCR β VDJ rearrangement sequence, a TCR β DJ rearrangement sequence, a TCR α VJ rearrangement sequence , A VCR rearrangement sequence of TCRγ, a VDJ rearrangement sequence of TCRδ, or a portion of a VD rearrangement sequence of TCRδ. In yet another embodiment, at least a portion of an IgH VDJ rearrangement sequence, an IgH DJ rearrangement sequence, and an IgK VJ rearrangement sequence is between the two binding positions of the first and second primer sets. Exists. And in other embodiments, between the two binding positions of the first and second primer sets, at least a TCR β VDJ rearrangement sequence, a TCR γ VJ rearrangement sequence, and a TCR δ VDJ rearrangement sequence. There is a component sequence, or part of the VCR rearrangement sequence of TCR δ. In some embodiments, at least the VDJ reconstruction portion comprises a complete D or NDN portion and V and J segment portions sufficient to identify them. In some embodiments, at least the VDJ reconstructed portion comprises a fragment of at least 50 nucleotides comprising a complete D or NDN portion and portions of the V and J segments. In some embodiments, at least the VDJ reconstruction portion comprises a fragment of at least 70 nucleotides comprising a complete D or NDN portion and portions of the V and J segments.

  In some embodiments, the first set includes one or more primers that are specific for J fragment or C fragment, respectively. After such a first set of primers anneals to their target sequences and extends, the non-extended primers in the first set are removed. A second set of primers, each specific for the V fragment, anneals to their target sequence and extends. In other embodiments, the first set includes primers specific for each V fragment, and after such first set of primers anneals and extends to their target sequence, the first set extends. Unprimed primers are removed and a second primer, which is specific for the J or C fragment, respectively, anneals to their target sequence and extends. In any alternative to these embodiments, the first and second sets may each contain a plurality of primers, and each primer may be specific for a different immunoreceptor fragment.

Returning to FIG. 1A, in some embodiments, the first and second sets of primers, in alternative embodiments, undergo melting, annealing, and extension cycles 1-10, or 2-10, or 3 After performing 10, or 4-10, or 5-10 cycles, extension is performed by removing unextended primers from the reaction mixture using conventional methods (5). In other embodiments, the first and second sets of primers, in alternative embodiments, are conventional after 2-5, or 3-5, or 4-5 cycles of melting, annealing, and extension cycles. Using the method, it is extended by removing the non-extended primer from the reaction mixture (5). In yet another embodiment, the first and second sets of primers are extended by performing two cycles of melting, annealing, and extension. For example, non-extended primers can be removed by exonuclease digestion, hybridization to complementary sequences on magnetic beads, or molecular sieve chromatography, commercially available spin columns (eg, Qiagen QIAquick PCR purification kit), etc. it can. In one embodiment, unextended or unincorporated primers are removed, for example, by digestion with exonuclease I. Double-stranded DNA (18), the product of extension (5), has a common first and second primer binding site at each end, and (in some embodiments) to these binding sites. Forward and reverse primers (6 and 11) with complementary sequences may be added to the subsequent cluster generation by bridge PCR. In some embodiments, the double stranded DNA also has a sequence tag (19) and the forward or reverse primer is used to identify or track or bind the sample or patient DNA (18). Tag (2) may be included. In some embodiments, the sequence tag (19) is substantially unique to each different reconstituted nucleic acid in the sample. As fully described below, the sequence tag (19) can be used to bind sequence reads to clonotypes and to detect and track sample contamination. For forward and reverse primers, primer binding sites (4) and (8) are used to perform bridge PCR (13) with certain sequencing protocols, eg, Genome Analyzer (Illumina, San Diego) (17). Can also be included. In other embodiments where one or more extensions are performed with a primer containing a sequence tag, each different reconstituted nucleic acid in the sample may have a copy with a different sequence tag attached thereto, Thus, for example, according to the embodiment described in FIG. 1A, the target polynucleotide is subjected to four cycles of melting, annealing, and extending, and the sample is reconstituted nucleic acid, S ₁ . If contained, upon completion of amplification (13) by common primer, a copy of the S ₁ has four different sequence tag at the maximum. Thus, sequence lead S ₁ can have four different sequence tag at the maximum. As will be fully explained below, in such an embodiment, the clonotype combines aligning sequence tags and combining sequence reads contained in each subset defined by a common sequence tag. Can be determined.

  In another embodiment, PCR is performed using common primers after at least two extension steps and two removal steps of unincorporated primers. As illustrated in FIG. 1B, primer (101) anneals to one end of a target polynucleotide (100), such as a reconstituted nucleic acid encoding an immunoreceptor chain, and is extended by, for example, a DNA polymerase. Primer (101) can include a receptor-specific portion (103) and a 5 ′ non-complementary portion (105), respectively, and similarly, sequence tag (104) and first primer binding site ( 102). As described above, after extension and removal of the unincorporated primer (130), the first extension product (109) in the reaction mixture is converted to (a) primer (125) [where each primer is a receptor A specific portion (106) and a 5′-non-complementary portion (115) (containing a primer binding site); and (b) a first primer binding site (102) and a 5 ′ non-complementary portion. A primer (127) comprising a portion (108) specific for (117) is added. Primers (125) and (127) are annealed to their primer binding sites and extended (107) to produce a second extension product (118), after which the non-extended primer is removed. A common forward primer (112) and reverse primer (110) are added to the second extension product (118), PCR is performed (111), and the resulting amplicon is sequenced (120 ). As in the embodiment of FIG. 1A above, whenever one or more extension steps are performed in the presence of a primer [(101) etc.] containing a sequence tag, the same target polynucleotide (100 ) May be labeled with a plurality of different sequence tags.

  FIG. 1C illustrates another embodiment using explicit V, D and J regions. Specific to primer (1212), ie V region (1226), against reconstituted nucleic acid (1200) encoding an immunoreceptor such as TCR in the reaction mixture under conditions that cause the primer to anneal Add first set of primers. Each first set of primers (1212) includes a receptor-specific portion and a 5′-non-complementary portion, and similarly, if desired, a sequence tag and a first primer binding site (eg, 1B includes 102, 103, and 104). The first set of primers (1212) anneals to the V region (1226) of the reconstituted nucleic acid (1200), and the first set of primers (1212) extends from the D region (1224) to at least the J region (1222). And optionally extending to the C region (1220) (1202) to produce a first extension product (1216) optionally comprising a sequence tag (1228) and a first primer binding site (1230). After removal of the first set of non-extended primers (1212), a second set of primers (1240) was added to the reaction mixture under annealing conditions to anneal to their respective target J regions (1222). The primers are then extended (1204) to produce a second extension product (1232) that includes a sequence tag (1236) (optional) and a second primer binding site (1234), respectively. The second extension product (1232) is, for example, adjacent to the V region (1226) as indicated by the sequence tag (1228) or adjacent to the J region (1222) as indicated by the sequence tag (1236). Alternatively, the second extension product (1232) may have two sequence tags placed on both parts. In one embodiment, the second extension product (1232) comprises a single sequence tag (1228) adjacent to the V region (1226).

  In another embodiment, the second extension product (1232) comprises a sequence tag (1236) adjacent to the J region (1222). In some embodiments, sequence tags (1228) and / or (1236) are mosaic tags as described below. After removing the unextended primer from the second set of primers (1240), common forward and reverse primers specific for the first and second primer binding sites (1230) and (1234), respectively, And PCR was performed (1206). The resulting amplicon sample is sequenced (1208) to generate sequence reads for constructing clonotypes and clonotype profiles.

  FIG. 1D illustrates a method for generating a template of a predetermined length and a method for adding one or two sequence tags to the template. Although the embodiment of FIG. 1D shows messenger RNA (mRNA) as a starting material, either DNA or RNA samples may be used in the method. One or more primers (1312) specific for the C region (1308) (“C primer”) are annealed to mRNA (1300) containing the VDJ region. Usually only a single C primer is used. Alternatively, one or more primers specific to the J region (having a similar structure) may be used. The C primer (1312) comprises a target specific fragment (1313), a sequence tag fragment (1314), and a common primer binding site (1315). Similarly, the target mRNA (1300) is annealed with a polymerase blocker (1310), which can be specific for the V region (1302) oligonucleotide. In some embodiments, only if the polymerase used to extend primer (1312) does not have either strand displacement activity or 5 ′ → 3 ′ exonuclease activity, and the oligonucleotide is non-extensible. For example, the blocker (1310) may be a natural oligonucleotide only if it has 3′-dideoxynucleotides. Blockers (1310) are typically oligonucleotide analogues such as antisense compounds with enhanced binding activity and nuclease resistance. In some embodiments, the blocker (1310) can be a locked nucleic acid (LNA) or a peptide nucleic acid (PNA) or a cross-linked nucleic acid (BNA), these nucleic acids can be found in the following references, Wengel et al., U.S. Patent Nos. 6,794,499; 7,572,582; Vester et al., Biochemistry, 43 (42): 13233-13241 (2004); and Kazuyuki et al., Chem. Comm. , 3765-3767 (2007); Nielson et al, Chem. Soc. Rev. 26: 73-78 (1997). The sequence of the blocker (1310) is selected such that the extension of the primer (s) (1312) stops at a predetermined position on the V region (1302). In some embodiments, the blocker (1310) is designed such that only sufficient V region (1302) is replicated in the extension process and the V region can be distinguished from the replicated sequence. In some embodiments, the consensus sequence can be selected to allow some mismatch, so that it is not necessary to obtain a blocker (1310) specific for each V region as long as the polymerase progression is stopped. The length of the blocker (1310) can be varied depending on the type of oligonucleotide or analog used. In some embodiments, the blocker (1310) has a length in the range of 10-25 monomers. In some embodiments, the blocker (1310) can be annealed to different positions on different V region sequences.

  Returning to FIG. 1D, primer (1312) extends to blocker (1310) to generate a cDNA copy of the VDJ portion of target (1300) with a predetermined length. In some embodiments, the predetermined length (ie, corresponding to the binding site of blocker (1310)) covers the desired portion of the VDJ region with one or more sequence reads of the sequencing method used in the method. Selected to get. After extension is complete, the RNA template (1300) is digested (1325) using conventional methods, for example, RNase (such as ribonuclease H and / or ribonuclease A) to obtain single stranded cDNA (1326). This cDNA is appended with a 3 'mononucleotide tail, eg, a polyC tail, using terminal deoxynucleotide transferase (TdT) in a general protocol. After overhanging the complementary adapter (1336) to the mononucleotide tail of the cDNA (1331) to which the tail was added, this adapter was extended to generate a double-stranded DNA (1340). Strand DNA can be amplified, for example, by PCR (1337), and the resulting amplicon sequenced (1338).

  Reconstituted nucleic acids that undergo hypermutation, such as nucleic acids encoding IgH, can be amplified using primer sets that include primers that bind to different primer binding sites on the same reconstituted nucleic acid. That is, such a set can include primers that bind to one or more non-overlapping primer binding sites on the same reconstituted nucleic acid encoding the receptor strand. Such a set may include either or both of a first set of primers and a second set of primers. In some embodiments, the reconstituted nucleic acid undergoing hypermutation is amplified with a first set of primers and a second set of primers, wherein at least one of the two sets does not overlap. Includes primers specific for a plurality of primer binding sites, eg, one set contains different V fragments, and the plurality of primers bind to different non-overlapping primers that bind to different V fragments. It is specific. Embodiments applicable to the amplification of reconstituted nucleic acids that undergo hypermutation are shown in FIGS. In the figure, the nested primer set is used to reliably amplify different reconstituted nucleic acids in a sample under conditions such as somatic hypermutation or clonal evolution. A reconstituted nucleic acid (eg, a molecule encoding IgH) is subjected to a first nested set of primers (302) under annealing conditions [in this example, the V region of reconstituted nucleic acid (300) ( 316) including groups specific to different sites (304), (306) and (308)]. In this embodiment, the first nested set includes a plurality of primer groups that are specific for different sites or positions in the V region, and different members in the group are specific for V region variants with different sites. . In some embodiments, the plurality of groups is 2-4, and in other embodiments, the plurality is 2 or 3. In some embodiments, each primer of the first nested set (302) has a first primer binding site (312) in the 5 'non-complementary tail in addition to a unique sequence tag (314). obtain. The primers of the first nested set (302) anneal to their target reconstituted nucleic acids and extend over the D region (318) and at least a portion of the J region (320) to form a first An amplicon (323) is generated. This first amplicon includes three products (330), (332) and (334) corresponding to three subprimer sets (304), (306) and (308), respectively. Each member of the first amplicon (323) incorporates a sequence tag (324) and a primer binding site (326).

  After removal of non-extended primer (322), a second nested set of primers (340) is added to the reaction mixture under annealing conditions. As illustrated in FIG. 3A, the second nested set of primers (340) includes primer subsets (336) and (338), which are the J region of the first amplicon (323) member. (320) Anneal to different non-overlapping parts. In some embodiments, the second nested set of primers may contain only a single primer group. The primers of the second nested set (340) are extended to produce a second extension product (360) comprising subsets (350), (352) and (354), each of which is now , Two more subsets (sub-subsets) corresponding to primers (336) and (338). In some embodiments, the second nested set of primers (340) contains primers specific for only a single primer binding site and the first nested set of primers (302) overlaps. Contains primers specific for at least two primer binding sites. After removal of non-extended primers (342), common forward and reverse primers can be added to perform PCR (356) and the resulting amplicon sample can be sequenced (358). In various embodiments, both the first nested set primer and the second nested set primer may contain a sequence tag (339); the first nested set primer contains a sequence tag; The nested set may not contain a sequence tag; and the second nested set of primers may contain a sequence tag and the first nested set may not contain a sequence tag. In some embodiments, the first nested set of primers is first extended, then the non-extended primers are removed or destroyed, and the second nested set of primers is annealed and extended (FIGS. 3A-3B). As shown). In other embodiments, the order of the annealing step, the extension step, and the removal step is reversed, i.e., first extending the primer of the second nested set, then removing or destroying the non-extended primer, Annealing and extending a set of nested primers.

  In some embodiments of the above methods, either one or more of the extension steps (322) or (342) is performed, eg, for the purpose of adding a sequence tag to the majority of the target polynucleotide in the sample. Is done. In such embodiments, one or more different sequence tags can be added to the target polynucleotide and / or copies thereof. That is, target polynucleotides and their replicating molecules derived from amplification reactions such as PCR can be tagged with multiple different sequence tags, so that a copy of the original target polynucleotide is made up of one or more sequence tags. Can be labeled. As will be explained fully below, such multiple sequence tags are still useful for tracking carryover contamination and more sensitively determining target polynucleotide sequences.

  Some of the above embodiments can also be implemented using the following steps. For example, a method for generating a clonotype profile from multiple or multiple T cell receptor chains comprises the following steps: (a) Regenerating from a first set of primers and T cells in a reaction mixture under primer extension conditions. Combining the configured nucleic acid sample [wherein each primer of the first set has a receptor-specific portion, wherein the receptor-specific portion is a predetermined number of target reconstituted nucleic acids. Each of the first set of primers comprises a length that anneals to a different reconstituted nucleic acid at a position or site and extends to generate a first extension product, each primer containing a first primer binding site. Having non-complementary ends]; (b) removing a first set of non-extended primers from the reaction mixture; (c) a second in the reaction mixture under primer extension conditions. Adding a primer set [wherein each primer of the second set has a receptor-specific portion, wherein the receptor-specific portion is added to the first extension product at a predetermined position or site. Annealing and having a 5′-non-complementary end containing a second primer binding site, wherein the first set of primers and / or the second set of primers are respectively a receptor-specific portion and a first or A sequence tag disposed between the second primer binding site and each second set of primers is extended to produce a second extension product, wherein each second extension product is A first primer binding site, a second primer binding site, at least one sequence tag, (i) a Vβ segment portion and a Jβ segment portion of a T cell receptor chain, (ii) T cell reception A Vδ segment portion and a Jδ segment portion of the chain, or (iii) a Vγ segment portion and a Jγ segment portion of the T cell receptor chain]; (d) performing a polymerase chain reaction in the reaction mixture, Generating a recon [this polymerase chain reaction uses a forward primer specific for the first primer binding site and a reverse primer specific for the second primer binding site]; and (e) Sequencing the amplicon nucleic acid to generate a clonotype profile of multiple T cell receptor chains. As used herein, “primer extension conditions” in a reaction mixture include conditions such that substantially all primer binding sites are in a single-stranded state. In some embodiments, such conditions are obtained by melting double stranded target nucleic acids, which results in the primer binding sites becoming single stranded and primer primers Can be annealed to form a foothold for extension by the polymerase.

  The predetermined positions or sites to be bound to the first and second sets of primers are determined by conventional methods known by practitioners of multiplex nucleic acid amplification methods (such as multiplex PCR) as exemplified by the references cited below. Also good. For example, if the target polynucleotide is a reconstituted nucleic acid encoding an immunoreceptor molecule, Faham and Willis (cited above), Van Dongen et al, Leukemia, 17: 2257-2317 (2003) and the like References provide guidance for selecting primer binding sites in multiplex amplification of target polynucleotides. In some embodiments, the selection of such a predetermined location or site is (i) an effect on amplification efficiency (the frequency of different copies in a faithful amplicon represents the frequency of the target polynucleotide in the sample). (Ii) the effect on the length required for the copy in the amplicon in the chemical procedure utilized for DNA sequencing; (iii) the selected primers have the desired diversity It is based on several factors, such as a part of the reconstituted nucleic acid (eg spanning the VDJ region). For this embodiment, the present invention reads out the sequence set rather than the analog signal, so that the cross-reactivity of the primers with different target polynucleotides does not affect the results of the methods of the present invention (eg, PCR Identification and recognition of amplification products and spectra (not given compared to methods based solely on analog readout, such as spectral input).

  In some embodiments, the sequencing step comprises the steps of: (i) providing a plurality of sequence reads each having an error rate, each comprising a nucleotide sequence and a tag sequence; ) Aligning sequence read groups having similar tag sequences and then making a base call based on the group sequence reads to determine the nucleotide sequence. Such group level nucleotide sequences can then be linked to the same or different clonotypes as described below. In some embodiments, in the PCR step, the length of the receptor-specific portion of the first and second sets of primers is such that the relative levels of different reconstituted nucleic acids in the amplicon are It is selected to be substantially the same as the level of the constructed nucleic acid. Thus, in selecting a primer, you may change the position and length of the binding site of the primer with respect to each target polynucleotide. In some embodiments, the sequence tag is selected from the set of sequence tags such that it is sufficiently larger than the number of different target polynucleotides in the sample, so that substantially all distinct target polynucleotides in the sample are Nucleotides and their copies will have different sequence tags (eg by the method described in Brenner, US Pat. No. 7,537,897, “labeling by sampling”). In some embodiments, the number of sequence tags in such a set is at least 100 times the total number of target polynucleotide populations in the sample. Further, in some embodiments where substantially all of the original target polynucleotide and their copies are labeled with the same unique sequence tag, the sequencing step generates a sequence read of the amplicon nucleic acid. And aligning sequence reads having the same sequence tag to determine sequence reads corresponding to the same clonotype of the sample. Further, in some embodiments, the aligning step further comprises determining the nucleotide sequence of each clonotype by determining a major nucleotide at each nucleotide position of a sequence read having the same sequence tag. . Further, in some embodiments, removing the non-extended primer digests single stranded nucleic acids in the reaction mixture using a nuclease having 3 ′ → 5 ′ single stranded exonuclease activity. (Digestion can be performed, for example, with E. coli exonuclease I, which can be easily inactivated by heat). In further embodiments, the methods described above can be used to generate a chronotype profile for diagnosing and / or monitoring minimal residual disease in cancer patients such as myeloma, lymphoma, or leukemia patients. Such diagnosis and / or monitoring may be performed after the steps of the above method, from the following additional steps: from the presence of one or more patient-specific clonotypes associated with cancer, and / or from the clonotype profile It can also be implemented by determining the level. The method of this embodiment includes determining the sequence of each of the one or more sequence tags, and comparing such sequences to the sequence of the sequence tag of the clonotype profile determined so far, It may further comprise determining the presence, absence and / or level of the sequence. In some embodiments, such a comparison step includes the sequence of one or more sequence tags in a clonotype database containing clonotypes from at least one individual other than the patient. Comparing with sequence tags.

  In yet another embodiment, the method of amplifying a plurality of reconstituted nucleic acids encoding T cell receptor components β, δ and γ in a single reaction comprises the following steps: (a) reaction Combining a first set of primers with a T cell-derived reconstituted nucleic acid sample under primer extension conditions in a mixture [wherein the different reconstituted nucleic acids are at least at the T cell receptor at the first end. Wherein each primer of the first set has a receptor-specific portion, wherein the receptor-specific portion is the first of the different reconstituted nucleic acids. Each primer of the first set has a length that anneals to one end and extends to produce a first extension product, each primer in the 3 ′ → 5 ′ direction and a first primer binding site Containing 5'-non Having complementary ends and the sequence tag is substantially different for each first set of primers]; (b) removing the first set of non-extended primers from the reaction mixture; (c ) Adding a second primer set to the reaction mixture under primer extension conditions [each primer of the second set has a receptor-specific portion, wherein the receptor-specific portion is With a length such that the extension product is annealed and extended to produce a second extension product, each second extension product comprising at least a portion of a Vβ, Vδ, or Vγ fragment of the T cell receptor Each primer of the second set has a 5′-non-complementary end containing a second primer binding site]; and (d) a polymerase chain reaction is performed on the reaction mixture to obtain an amplicon. To produce the the polymerase chain reaction, a forward primer specific to the first primer binding site, using the reverse primers specific to the second primer binding sites] includes, a. The above method may further comprise the step of sequencing an amplicon sequence sample. Typically, such a sample is a “representative sample” and is large enough to have different clonotypes present in the sample at approximately the same frequency as in the original biological sample. In some embodiments, the sequencing step includes providing a plurality of sequence reads each having an error rate, each including a nucleotide sequence and a tag sequence, and aligning sequence reads having similar tag sequences. Determining sequence reads corresponding to the same clonotype. Whenever multiple sequence tags have been added to the original target polynucleotide or copies thereof, the sequence reads can be processed with the binding step described more fully below.

  In another embodiment, a method for generating a clonotype profile from multiple T cell receptor chains comprises the following steps: (a) from a first set of primers and T cells in a reaction mixture under primer extension conditions. Combining the reconstituted nucleic acid sample [where each first set of primers has a receptor-specific portion, wherein the receptor-specific portion is reconstituted differently at a given position. Annealing to the nucleic acid and extending to generate a first extension product, each first set of primers having a 5′-non-complementary end containing a first primer binding site]; (B) removing the first set of unextended primers from the reaction mixture; (c) adding a second set of primers to the reaction mixture [where the second set of primers] These primers have a receptor-specific portion with a certain length, the receptor-specific portion being specific for the first extension product at a predetermined position, and the second primer binding Having a 5'-non-complementary end containing a site, wherein the first set of primers and / or the second set of primers are each of a receptor-specific portion and a first or second primer binding site Including a sequence tag disposed between]; (d) performing a first polymerase chain reaction to produce a first amplicon [this first polymerase chain reaction comprises a first primer binding site; And a second set of primers, each nucleic acid sequence of each first amplicon comprising a first primer binding site, a second primer binding site, and at least one Including a row tag and either a Vβ segment portion and a Jβ fragment portion of a T cell receptor chain, a Vδ segment portion and a Jδ fragment of a T cell receptor chain, or a Vγ segment portion and a Jγ segment portion of a T cell receptor chain The length of the receptor-specific portion of the first and second sets of primers is such that, in the amplicon, the relative levels of the different reconstituted nucleic acids are substantially the same as the levels of the different reconstituted nucleic acids in the sample. (E) adding a reverse primer specific for the second primer binding site; (f) performing a second polymerase chain reaction on the reaction mixture to produce a second amplicon. [This polymerase chain reaction comprises a forward primer specific for the first primer binding site and a reverse direction specific for the second primer binding site. And (g) sequencing the nucleic acid of the second amplicon to generate a clonotype profile of the T cell receptor chain. In some embodiments, the sequencing step includes providing a plurality of sequence reads each having an error rate, each including a nucleotide sequence and a tag sequence, and aligning sequence reads having similar tag sequences. Determining sequence reads corresponding to the same clonotype. In further embodiments in which the target polynucleotide and / or copies thereof are labeled with one or more sequence tags, after alignment of similar sequence tags, the sequence read can be further coupled as described more fully below. Can also be processed.

  In another example, a method for generating a clonotype profile from multiple B cell receptor chains comprises the following steps: (a) a first nested set of primers under primer extension conditions in a reaction mixture, and B Combining the reconstituted nucleic acid sample from the cell, wherein the first nested set comprises one or more primer groups, each primer of each group having a receptor specific portion, Sometimes the receptor-specific portion has a length such that the receptor-specific portion of each primer from a different group anneals to a different reconstituted nucleic acid at a predetermined site, the predetermined site comprising: Each primer of each group does not overlap with a predetermined site of any other primer in the nested set, and the first primer Having a 5′-non-complementary end containing a binding site]; (b) extending a primer of the first nested set to generate a first extension product; (c) a first nested product; Removing unextended primers from the set reaction mixture; (d) adding a second nested set of primers to the reaction mixture under primer extension conditions [where the second nested set is One or more primer groups, each primer of each group having a receptor-specific portion, wherein the receptor-specific portion of each primer from a different group is defined at a predetermined site. It has a length that allows it to anneal to one extension product, and this predetermined site is over the predetermined site of any other primer in the second nested set. Each primer of each group has a 5′-non-complementary end containing a second primer binding site, and the first nested set of primers and / or the second nested set of The primer comprises a sequence tag disposed between the receptor-specific portion and the first or second primer binding site, respectively]; (e) a second extension by extending the primer of the second nested set; Generating a product [wherein each second extension product comprises a first primer binding site, a second primer binding site, at least one sequence tag, and (i) a B cell receptor weight chain V segment part and J segment part or (ii) any of V segment part and J segment part of B cell receptor kappa light chain]; (f) in the reaction mixture Performing a polymerase chain reaction to produce an amplicon [this polymerase chain reaction comprises a forward primer specific for the first primer binding site and a reverse primer specific for the second primer binding site; And (g) sequencing the amplicon nucleic acid to generate a chronotype profile of multiple B cell receptor chains.

  In some embodiments, one or more cycles of annealing and extending the primer (after melting of the extension product) can be performed in steps (b) and / or (e), in which case A copy of the original reconstituted nucleic acid may be labeled with one or more sequence tags. In these embodiments, the sequencing step (g) may further include an alignment and binding step as described below for determining clonotype and clonotype profiles. In some embodiments, for example, if only a single extension is made in steps (b) and (e), the sequencing steps each have an error rate and each comprises a nucleotide sequence and a tag sequence. Providing a plurality of sequence reads comprising, aligning sequence reads having similar tag sequences to determine sequence reads corresponding to the same clonotype. As described above, in some embodiments, in the PCR process, the position and length of the receptor-specific portion of the first and second sets of primers is the relative level of different reconstituted nucleic acids in the amplicon. Is selected to be substantially the same as that of the reconstituted nucleic acid in the sample.

  In some embodiments, the sequence tag is added to the target polynucleotide or a copy thereof in a primer extension step, and substantially all different target polynucleotides and their copies are labeled with the same sequence tag. In other embodiments, the target polynucleotides of the samples or their copies can be labeled with one or more different sequence tags. As further exemplified below, in some embodiments, multiple extensions or multiple times in the presence of a primer (either a first set of primers or a second set of primers) containing a sequence tag. PCR cycles can be performed, which can result in the addition of different sequence tags to the same target polynucleotide and / or copies thereof.

Sequence tags in clonotype analysis In one aspect, the present invention obtains and analyzes sequence data from repertoires of immune molecules such as T cell receptor (TCR) or B cell receptor (BCR) or certain fragments thereof. It aims at a method for quickly and efficiently determining a chronotype profile. The sequence data typically includes a large collection of sequence reads, ie, the base call sequence from the DNA sequencer and the associated quality score are used to analyze the immune molecule. In constructing a chronotype profile, there are really differences from sequence reads that contain errors from non-biological sources such as extraction processes, chemical techniques for sequencing, or chemical techniques for performing amplification. It is an important issue to quickly and accurately identify the reading sequence reads. Aspects of the invention provide for each target polynucleotide (eg, a reconstituted nucleic acid in a sample) to assist in determining whether the sequence reads of the complex are derived from the same original target polynucleotide. , Including adding unique sequence tags. According to one aspect of the present invention, a sequence tag is added to a somatic reconstituted nucleic acid molecule to produce a tag-molecule complex, each reconstituted nucleic acid of such a complex being a unique sequence tag. Have Usually, such addition is done after extracting nucleic acid molecules from a sample containing T cells and / or B cells and / or cell-free DNA. Preferably, such unique sequence tags are as different as possible from each other as assessed by conventional sequence distance measurement methods such as Hamming distance. By maximizing the distance between sequence tags in a tag molecule complex, the sequence tag of the complex is better than any other tag sequence in a different complex, even when sequencing and amplification errors are high. Are also retained much closer to the original tag sequence. For example, utilizing a 16-mer sequence tag and each such tag in the set of clonotypes has a Hamming distance at a distance of at least 50% or 8 nucleotides from all other sequence tags of the clonotype In some cases, converting a tag to another and causing misreading of the sequence tag requires at least 8 sequencing or amplification errors (and inappropriate chronotype sequence reads with incorrect sequence tags) Need to be grouped). In one embodiment, the sequence tag is selected after adding to the reconstituted nucleic acid molecule to produce a tag-molecule complex, and the Hamming distance between the tags of the tag-molecule complex is at least the full length of the sequence tag. A numerical value of 25% (ie, each sequence tag is at least 25% nucleotides different from the sequence of all other tags); in another embodiment, humming between such sequence tags The distance is a value that is at least 50% of the total length of the sequence tag.

  In one aspect, the present invention provides the following steps: (a) obtaining a sample comprising T cells and / or B cells and / or cell-free DNA from an individual; (b) a T cell receptor gene in the sample. Or adding a sequence tag to the reconstituted nucleic acid molecule of an immunoglobulin gene to produce a tag-molecule complex (where substantially all tag-molecule complexes have unique sequence tags) (C) amplifying the tag-molecule complex; (d) sequencing the tag-molecule complex; and (e) aligning sequence reads of similar sequence tags to re-same the same in the sample. Determining a sequence read corresponding to the constructed nucleic acid. Samples containing B cells or T cells are obtained using conventional methods. In the step of adding a sequence tag, preferably the sequence tag is not only unique but also sufficiently different from another tag, i.e. even if the sequencing or amplification error is quite large. The likelihood of being changed to a tag of is close to zero. Although most sequencing techniques require amplification of the tag-molecule complex after the addition of a sequence tag, the amplification step is optional when using single molecule sequencing methods. Single molecule sequencing methods include, but are not limited to, single molecule real time (SMRT) sequencing, or nanopore sequencing, such as US Pat. No. 7,313,308; No. 375; No. 7,907,800; No. 7,960,116; No. 8,137,569; and Manrao et al, Nature Biotechnology, 4 (8): 2685-2893 (2012). Can be mentioned.

  In another aspect, the invention includes a method for determining the number of lymphocytes in a sample by counting unique sequence tags. Without using sequence tags, the clonotype of the TCRβ or IgH gene, particularly the clonotype containing the V (D) J region, provides a unique marker for lymphocytes and their clones. If the reconstituted nucleic acid is obtained from genomic DNA, the number of lymphocytes in the sample can also be assessed by the number of unique clonotypes counted after sequencing. This approach is used whenever there is a substantially clonal population of identical lymphocytes associated with the same clonotype (or the reconstituted nucleic acid is derived from the sample mRNA and the quality of each sequence is May reflect the expression rate as well as the number of cells, or will respond). The use of sequence tags overcomes this short coming and is particularly useful for counting lymphocytes in patients suffering from various lymphatic diseases such as lymphoma or leukemia. According to one aspect of the invention, sequence tags can also be used to obtain the absolute number of lymphocytes in a sample regardless of whether a large number of clones such as leukemia are present. Such a method comprises the steps of: (a) obtaining a sample containing lymphocytes from an individual; (b) adding a sequence tag to the reconstructed nucleic acid molecule of the T cell receptor gene or immunoglobulin gene of the lymphocyte. Generating a tag-molecule complex (wherein tag-molecule complexes of substantially all molecules have unique sequence tags); (c) amplifying the tag-molecule complex; d) sequencing the tag-molecule complex; and (e) counting the number of different sequence tags to determine the number of lymphocytes in the sample. In some embodiments, the reconstituted nucleic acid molecule is derived from genomic DNA.

In one embodiment of the present invention, the sequence tag is, for example, Brenner et al., US Pat. No. 5,846,719; Brenner et al., US Pat. No. 7,537,897; Macevicz, International Publication No. WO 2005/111242. Are added to the reconstituted nucleic acid molecule of the sample by labeling by sampling, such as those disclosed in; and the like; these references are incorporated herein by reference. For sampling by labeling, a group of polynucleotides that are labeled (ie, appended with a unique tag) are used (by addition, binding, or similar methods) to a larger group of sample sequence tags. That is, if the polynucleotide population has K members (including duplicates of the same polynucleotide) and the sequence tag population includes N members, then N >> K. In one embodiment, the size of the sequence tag population used in the present invention is at least 10 times the size of the clonotype population in the sample; in another embodiment, the size of the sequence tag population used in the present invention. Is at least 100 times the size of the clonotype population in the sample; in another embodiment, the size of the sequence tag population used in the present invention is at least 1000 times the size of the clonotype population in the sample. Is double. In other embodiments, the size of the sequence tag group is such that when the clonotype is combined with the sequence tag group, whenever it is combined in an additional reaction, such as, for example, a ligation reaction or an amplification reaction, In general, all clonotypes are selected to have a unique sequence tag. In some embodiments, substantially all chronotypes mean that at least 90% of such chronotypes may have unique sequence tags; in other embodiments, substantially all Means that at least 99% of such chronotypes may have unique sequence tags; in other embodiments, substantially all chronotypes of such chronotypes This means that at least 99.9% may have a unique sequence tag. In many tissue samples or biopsies, the number of T cells or B cells is up to or about 1 × 10 ⁶ ; thus, in some embodiments of the invention using such samples, labeling by sampling The number of unique sequence tags used in is at least 10 ⁸ , or in other embodiments at least 10 ⁹ .

In such an embodiment, up to 1 × 10 ⁶ clonotypes are labeled by sampling and the sequence is obtained by combinatorial synthesis by reacting all four nucleotide precursor mixtures in each addition step of the synthesis reaction. Large sets of tags can be generated efficiently (eg, as disclosed in Church, US Pat. No. 5,149,625, incorporated by reference). As a result, the set of sequence tags will have the structure “N ₁ N ₂ ... N _k ”, where each Ni = A, C, G, or T, and k is the nucleotide in the tag Is a number. The number of sequence tags in the sequence tag set produced by such combinatorial synthesis is 4 ^k . Thus, to add a sequence tag to a population of 10 ⁶ molecules by labeling by sampling, a sequence tag set in which k is at least 14 or k is in the range of about 14-18 is suitable. A sequence tag set having the above structure comprises a number of sequences that can introduce differences or errors during the performance of the method of the invention. For example, the combinatorially synthesized sequence tag set includes a number of member tags, including homopolymer fragments, in which case some sequencing approaches such as sequencing-by-synthesis If it exceeds a certain length, it becomes difficult to measure accurately. Thus, the present invention includes combinatorially synthesized sequence tags that have an efficient structure for certain method steps such as sequencing. For example, by splitting the four naturally occurring nucleotides into discrete subsets that are alternatively used in combinatorial synthesis to create several sequence tag structures that are efficient for synthesis decoding, homopolymers It is possible to prevent the predetermined length from being exceeded. For example, by setting z to either A or C and x to either G or T,
[(Z) ₁ (z) ₂ . . . (Z) _i ] [(x) ₁ (x) ₂ . . . (X) _j ]. . . Is obtained,
Here, i and j may be the same or different and are selected to limit the size of any homopolymer fragment. In one embodiment, i and j are in the range of 1-6. In such embodiments, sequence tags may have a length in the range of 12-36 nucleotides; and in other embodiments, such sequence tags have a length in the range of 12-24 nucleotides. May be included. In other embodiments, other nucleotides can be used as pairs, for example, z can be A or T and x can be G or C; or z can be A or G and x can be It can be T or C. Alternatively, if z ′ is a combination of three of the four natural nucleotides and x ′ is any nucleotide other than z ′ (eg, z ′ is A, C or G, then x ′ becomes T ). This gives the following sequence tag structure:
[(Z ′) ₁ (z ′) ₂ . . . (Z ′) _i ] x ′ [(z ′) ₁ (z ′) ₂ . . . (Z ′) _i ] x ′. . .
Where i is selected as described above, and the occurrence of x ′ is provided as a break, terminating any undesirable homopolymer.

Additional Sequence Tags The present invention uses a method of labeling nucleic acids, such as genomic DNA fragments, with a unique sequence tag that can include a “mosaic tag” prior to amplification and sequencing. Such sequence tags are useful for identifying amplification errors and sequencing errors. Mosaic tags minimize the sequencing and amplification of artifacts due to inappropriate annealing, priming, or hairpin formation, etc. that can be caused by conventional fully random sequence tags. In one aspect, the mosaic tag is a sequence tag comprising alternating constant and variable regions, each constant region having a position in the mosaic tag and comprising a predetermined sequence of nucleotides, each variable region comprising: It has a position in the mosaic tag and contains a predetermined number of randomly selected nucleotides. For illustration purposes, the 22-mer mosaic tag (SEQ ID NO: 1) may have the following form:

Region position In this case, there are a total of nine constant regions and variable regions. As variable regions (nucleotides with double underlining), regions 1 (nucleotides 1 to 3), 3 (nucleotides 9) and 5 (nucleotides) 12-14), 7 (nucleotides 18-19) and 9 (nucleotides 21-22), and the constant regions are regions 2 (nucleotides 4-8), 4 (nucleotides 10-11), 6 (nucleotides 15-17) ), And 8 (nucleotide 20). N represents a randomly selected nucleotide from the set of A, C, G, or T, so the number of mosaic tags in this example is 4 ¹¹ = 4,194,304 tags. b represents a predetermined nucleotide at the described position. In some embodiments, the sequence of b, ^{"*** bbbb * bb *** bbb ** b} ** ", compared the genome of the organism which constitutes the sample to minimize the likelihood of having an exact match Selected to suppress.

  With respect to the mosaic tag of a specific embodiment of the method of the present invention, in one aspect, all constant regions at the same location have the same length, and all variable regions at the same location have the same length. Thereby, it is possible to synthesize a mosaic tag using a general chemical reaction and apparatus and utilizing partially combinatorial synthesis.

  In one aspect, the mosaic tag comprises 10 to 100 nucleotides, or 12 to 80 nucleotides, or 15 to 60 nucleotides. In some embodiments, the mosaic tag comprises at least 8 nucleotide portions of randomly selected nucleotides; in other embodiments, the mosaic tag is at least 15 bases long nucleotides, of the nucleotide portion Nucleotides are randomly selected for at least 12 bases. In another embodiment, the mosaic tag has no variable region and is longer than 7 nucleotides in length.

In another embodiment, the mosaic tag is labeled by the following steps: (i) preparing a DNA template from the nucleic acid in the sample; (ii) sampling the DNA template to form a multiple tag-template complex. Wherein substantially all DNA templates of the tag-template complex have a unique mosaic tag comprising alternating constant and variable regions, each constant region having a predetermined sequence of mosaics. The position in the tag has nucleotides with a length of 1 to 10 bases, and each variable region has a position in the mosaic tag and a randomly selected nucleotide with a length of 1 to 10 bases , Constant regions at the same position have the same length, and variable regions at the same position have the same length; (iii) tag-temp (Iv) generating a plurality of sequence reads of each amplified tag-template complex; (v) a plurality of sequences having the same mosaic tag; Determining the nucleotide sequence of each nucleic acid by determining the matching nucleotide at each nucleotide position of the lead. In another embodiment, the mosaic tag comprises the following steps: (a) preparing a single stranded nucleic acid DNA template from the nucleic acid in the sample; (b) sampling the single stranded nucleic acid DNA template to form a tag-template complex. Labeling by forming, wherein substantially all of the single-stranded nucleic acid DNA template of the tag-template complex has a nucleotide length of at least 15 bases and has the following form:
[(N ₁ N ₂ ... N _Kj ) (b ₁ b ₂ ... B _Lj )] _M
With a unique sequence tag (ie, mosaic tag), where i = 1, 2,. . . K _j the respective _{N i} is the, A, C, a nucleotide that is randomly selected from the group consisting of G and T; _{K j} is an integer ranging from 1 to 10, each j is below M Yes (ie, region N ₁ N ₂ ... N _Kj is a variable region); i = 1, 2,. . . Each b ₁ that is L _j is a nucleotide; L _j is an integer in the range of 1-10, each j is less than or equal to M, and all sequence tags (i) have the same Kj for all j And (ii) the same array b ₁ b ₂ . . . has a b _Lj (i.e., the area _b 1 _b 2 ... _{b Lj} is constant region); and M is an integer greater than or equal to 2; (c) the tag - step of amplifying the template complex; (d (B) generating a plurality of sequence reads of each amplified tag-template complex; (e) determining each corresponding nucleic acid by determining the corresponding nucleotide at each nucleic acid position of the plurality of sequence reads having the same sequence tag. Can be used in the step of determining the nucleotide sequence. In some embodiments, there are at least 10 ⁴ sequence reads; in other embodiments, there are at least 10 ⁵ sequence reads; in still other embodiments, at least 10 ⁶ sequence reads. is there. In some embodiments, the total length of the sequence tag ranges from 15 bases to 80 bases.

Addition of Sequence Tags A variety of different addition reactions can be used to add unique tags to virtually all clonotypes in a sample in addition to the above. In order to add a sequence tag to the reconstituted nucleic acid, the routine modifications of the present invention include a subset of sample nucleic acids, eg, to reduce sample complexity in microarray or genomic sequencing methods. Many techniques for capturing can be used. Examples of techniques for capturing various sets of targeted nucleic acids, including sequence tag addition, sequencing, and similar methods for subsequent manipulations include Willis et al., US Pat. No. 7,700,323; Jones et al., US Patent Application Publication No. 2005/0142577; Gullberg et al., US Patent Application Publication No. 2005/0037356; Porreca et al., Nature Methods, 4 (11): 931-936 (2007); Turner et al, Nature Methods. 6 (5): 315-316 (2009); Church, US Pat. No. 5,149,625; and Macevicz, US Pat. No. 8,137,936.

  In one embodiment, such addition involves combining a sample containing a reconstituted nucleic acid molecule (also containing a clonotype sequence) with a group of sequence tags or a library of sequence tags. Groups of molecules can be randomly combined and associated or bound (eg, covalently bonded). For example, such a random combination can occur in a biomolecular reaction that anneals and extends a tag-containing primer to a target nucleic acid, or ligates an adapter containing a tag to the end of a target nucleic acid. In some embodiments, the method of adding a tag may depend to some extent on the DNA sequencing approach. For example, in sequencing methods that generate relatively long and accurate sequence reads, such as the 454 sequencing method, see, for example, 5′-RACE disclosed in Freeman et al, Genome Research, 19: 1817-1824 (2009). The sequence tag can be added by ligating an adapter containing a sequence-tag to one end or both ends after preparing a cDNA library from mRNA containing the reconstructed nucleic acid using a conventional method such as. In other embodiments, when using sequencing methods such as “Illumina” sequencing or “Ion Torrent” sequencing that produces relatively short and error-prone sequence reads, Additional steps may be required to have a length covered by the sequence reads generated by the technique. In such a tag addition reaction, the clonotype sequence includes a linear single-stranded or double-stranded polynucleotide, and the sequence tag includes an amplification primer such as a PCR primer, a ligation adapter, a circularizable probe, and a plasmid. Or can be transported by a reagent such as an equivalent. Some such reagents that allow for transport of sequence tags are Macevicz, US Pat. No. 8,137,936; Faham et al, US Pat. No. 7,862,999; Landegren et al, US Pat. 8, 053, 188; Unrau and Deugau, Gene, 145: 163-169 (1994); and Church, US Pat. No. 5,149,625, which are hereby incorporated by reference.

2A and 2B illustrate an addition reaction involving PCR. Sequence tag group _{(Τ 1, T 2, T} 3 ... T j, T j + i ... T k, T k + 1 ... T n-1, T n) are incorporated into primer (2100). The sequence tag group has a size sufficiently larger than the reconstructed nucleic acid molecule (2102). The sequence tag is added to the reconstituted nucleic acid molecule by annealing the primer to the nucleic acid molecule and extending the primer with DNA polymerase in the first cycle of PCR. The figure shows a method for sorting the reconstituted nucleic acid molecules, ie, based on a common primer binding region (2104), eg, V region (2108), a small fraction of the total population of sequence tags, The example of the method by annealing a primer at random is shown. Primers (and thus sequence tags) are randomly matched to the reconstructed nucleic acid sequence molecule, so there is a slight possibility that the same sequence tag will be added to different nucleic acid molecules, but the sequence tag group If as large as taught in the specification, this possibility is reduced to a negligible extent, and substantially every nucleic acid molecule has a unique sequence tag added. The other primer (2106) of the forward and reverse primer pairs anneals to the C region (2110), forming an amplicon (2112) after multiple annealing, extension, and melting cycles, , A unique sequence tag is added to the V (D) J region containing the group of chronotypes (2102). That is, the amplicon (2112) contains a tag-molecule complex derived from an addition reaction.

2C and 2D illustrate a method of adding a pair of sequence tags to each or substantially different reconstituted nucleic acid in a sample. Similar to the method of FIGS. 2A and 2B, the sequence tags (Τ ₁ , T ₂ , T ₃ ... T _j , T _{j + i} ... T _k , T _{k + 1} ... T _n−1 , T _n ) The transporting primer (2200) is used as a downstream primer, and the common primer (2106) is further used as a sequence tag (Tm, Tm + 1, Tm + 2 ... Tq, Tq + 1, Tq + 2, ... Tr, Tr + 1, Tr + 2, ...). .) (Ts, Ts + 1, Ts + 2,...) Are used as upstream primers. As with the downstream primer set, the number of different sequence tags transported by the upstream primer (2206) is such that the nucleic acid reconstituted so that substantially all reconstituted nucleic acid (2202) has a unique tag after amplification. It may be larger than the number of molecules (2202). In some embodiments, the respective set of sequence tags in primers (2206) and (2200) need not be as large as the set of sequence tags in the embodiment of FIGS. 2A and 2B. Sequence tags that label a single reconstituted nucleic acid even when different reconstituted nucleic acids share one sequence tag of the pair, since different reconstituted nucleic acids are uniquely labeled by the sequence tag pair The substantial uniqueness of the pair is not impaired. Thus, in the embodiment of FIGS. 2C and 2D, each primer set (2200) and (2206) has fewer types of sequence tags than the primer set (2100). For example, when a random sequence tag is used and a primer (2100) contains a 16-mer sequence tag, the primers (2200) and (2206) each contain an 8-mer sequence tag, and the sequence tag The overall diversity can be made comparable. Alternatively, the embodiment of FIGS. 2C and 2D is implemented similarly to FIGS. 2A and 2B. A sequence tag is added to the reconstituted nucleic acid molecule by annealing the primer to the nucleic acid molecule and extending the primer with DNA polymerase in the first cycle of PCR. As described above, FIG. 2C shows a method for sorting the reconstituted nucleic acid molecules, ie, having common primer binding sites (2204) and (2205), eg, V region (2208) and C region (2210), respectively. And an example of a method by annealing a small fraction of the entire population of sequence tag pairs to a primer at random. Since primers (and thus sequence tags) are randomly matched to the reconstituted nucleic acid sequence molecule, there is a slight possibility that the same sequence tag pair will be added to different nucleic acid molecules, but the sequence tag group is here If as large as taught in the document, this possibility is reduced to a negligible extent and a unique sequence tag is added to virtually all reconstituted nucleic acid molecules. . After multiple annealing, extension, and melting cycles, an amplicon (2212) is formed, which adds a unique sequence tag to the V (D) J region containing the group clonotype (2202). . That is, the amplicon (2212) contains a tag-molecule complex derived from the addition reaction.

  In some embodiments, for example, Porreca et al (supra); Willis et al (supra); or similar routine modifications disclosed by similar references capture the sequence tag and A circularizable probe can be used to add to the reconstituted nucleic acid. As shown in FIGS. 2E and 2F, the following components: upstream target binding fragment (2304); downstream target binding fragment (2306) with 5′-phosphorylated terminus (2305); sequence tag (2310); second A circularizable probe (2302) is provided that comprises a common primer binding site (2314); an optional cleavage site (2308); and a first common primer binding site (2312). Under annealing conditions, a circularizable probe (2302) contains the target polynucleotide (eg, may be single-stranded or double-stranded cDNA prepared from mRNA by common techniques) Combine with reaction mixture containing sample (2300). As shown, the target polynucleotide comprises the V, NDN, J and C regions of the reconstituted nucleic acid encoding IgH or TCRβ chain. In some embodiments, the sequences of the upstream and downstream target binding fragments (2304) and (2306) are each selected to span a portion of the VDJ region of the target polynucleotide. The circularizable probe (2302) and target polynucleotide (2300) form a complex (2330) in the reaction mixture by annealing the upstream and downstream target binding fragments (2304 and 2306). Upstream target binding fragment (2304) in the presence of DNA polymerase and dNTPs to downstream target binding fragment (2306) replicating (and thereby capturing) part of the VDJ region of the target polynucleotide Elongate (2340). The extended upstream target binding fragment is ligated to the downstream target binding fragment (2306) in the presence of ligase activity to form a closed single-stranded DNA circle (2342). Next, optionally, the reaction mixture may be treated (2344) with exonuclease to remove unreacted probe and target polynucleotide. In some embodiments, after cleaving the cleavage site (2308) and opening the single-stranded nucleic acid circle (2342) (this cleavage site is, for example, a recognition site for an endonuclease that cleaves at a rare position). Or a RNA monomer is inserted into the probe, cleaved with RNase H or the like, and then the VDJ-tag insertion part of the opened probe (2348) is amplified with primers (2350) and (2352). You may let them. Primers (2350) and (2352) may contain non-complementary regions to add elements that allow for subsequent DNA sequencing (2354). Alternatively, using single stranded nucleic acid circles for direct sequencing methods such as Drmanac et al, Science, 327 (5961): 78-81 (2010); and US Pat. No. 8,445,196, etc. A nanoball template may be generated.

  FIG. 2G illustrates another embodiment for adding a sequence tag to a reconstituted nucleic acid encoding an immunoreceptor molecule. Guidance in practicing this embodiment can be found in Faham and Zheng, US Pat. No. 7,208,295, incorporated herein by reference. In the reaction mixture, reconstituted nucleic acid (2450) is combined with probe (2454) and adapter (2456) under annealing conditions. Probe (2454) includes a receptor specific portion (2455) and an adapter specific portion (2457). For example, probe (2454) may comprise a probe mixture, where different probes have receptor-specific portions that are specific for different J regions, or in other embodiments, in different V regions. It has a specific, receptor-specific part. The 5'-phosphorylated adapter (2456) includes a probe-specific portion (2458), a sequence tag (2460), and a first primer binding site (2462) at the 5 'end. The position, sequence, and length of the receptor-specific portion (2455) and adapter-specific portion (2457) and probe-specific portion (2458) of probe (2454) are such that they hybridize to each other. To select the structure (2452). After formation of the structure (2452), the single-stranded nucleic acid portion (2461) is cleaved from the reconstructed nucleic acid (2450), and the cleaved 3 ′ end of the reconstituted nucleic acid (2450) is attached to the adapter (2456). After ligation with the 5 ′ phosphorylated terminus to form the first extension product (2459), the probe (2454) is removed (2474). Cleavage of (2461) can also be performed with single stranded nucleic acid nucleases as described by Faham and Zheng. In one embodiment, for example, as described by Faham et al, US Pat. No. 7,208,295, which is hereby incorporated by reference, for example, in a PCR where dTTp is replaced with dUTP, thymidine is replaced with uracil. Probe (2454) is synthesized and treated with uracil-DNA glycosylase (UDG) to remove the probe. By UDG treatment, the probe (2454) is cleaved with uracil to obtain a fragment (2455). After cleaving the probe, the adapter and the overhang are removed (2476), the forward primer (2466) and the reverse primer (2468) are added to the extension product (2464), PCR is performed (2470), and the resulting amplifier The recon sample is sequenced (2472).

  In an embodiment similar to FIG. 2G, a similar probe and adapter can be used to add a sequence tag to a predetermined site of the targeted polynucleotide. Here, a flap endonuclease such as FEN-1 is used for the cleavage of the single-stranded nucleic acid moiety corresponding to (2461). In this embodiment, if further different nucleases are used, multiple probe and adapter sequences are reversed; that is, the flap endonuclease substrate contains an adapter 3 ′ end (2454) corresponding to annealing to the target sequence (2450). ) And a single-stranded nucleic acid portion corresponding to the 5 ′ end (2452) of the target sequence. Subsequent steps are performed substantially simultaneously after cleavage and removal of the probe sequence. Guidance for utilizing flap endonucleases in detection assays can be found in the following references: Lyamichev et al, Nature Biotechnology, 17: 292-296 (1999); and Eis et al, Nature Biotechnology, 19: 673-676 (2001).

  In some embodiments, the reconstituted nucleic acid encodes an immunoreceptor molecule chain, which is typically a very large scale of very similar polynucleotides (eg, greater than 1000, The molecular chain may have a length of less than 500 bases of nucleotides, and the molecular chain may have a length of less than 500 bases. Or, in other embodiments, it may have a length of less than 400 base nucleotides, or in still other embodiments, less than 300 base nucleotides. In one aspect of the present invention, we find that these characteristics make use of very different sequence tags and efficiently compare very similar chronotype sequence reads so that they are in the same sequence. We recognize and understand that it was possible to determine whether it came from.

sample:
The term “sample” refers to a quantity of biomaterial, and in some embodiments, the biomaterial is obtained from a patient and contains cellular and / or cell-free DNA; Is also referred to as “specimen” or “tissue sample”. The term “sample” may be used from a statistical point of view, for example, in obtaining large sets or large subsets or portions of reconstituted nucleic acids; Can be understood to mean “representative specimens”, and such specimens are understood to reflect or approximate the relative frequency of different nucleic acids in the tissue (for example). One skilled in the art will recognize the appropriate usage of this term from context.

  The clonotype profile can be obtained from a body fluid sample such as blood containing immune cells or cell-free nucleic acids encoding immunoreceptor chains. For example, immune cells can include T cells and / or B cells. Examples of T cells (T lymphocytes) include cells expressing a T cell receptor. T cells include helper T cells (effector T cells or Th cells), cytotoxic T cells (CTL), memory T cells, and regulatory T cells. In one aspect, the T cell sample contains at least 1,000 T cells; more typically the sample contains at least 10,000 T cells, and more typically at least 10,000 T cells. 100,000 T cells are contained. In another embodiment, the sample contains a large number of T cells in the range of 1000 to 1,000,000. The immune cell sample can also include B cells. Examples of B cells include plasma B cells, memory B cells, B1 cells, B2 cells, marginal zone B cells, and follicular B cells. B cells can express immunoglobulins (antibodies, B cell receptors). In one aspect as described above, the B cell sample contains at least 1,000 B cells; more typically, the sample contains at least 10,000 B cells, and more typically Contains at least 100,000 B cells. In another embodiment, the sample contains a large number of B cells in the range of 1000 to 1,000,000.

  Samples used in the methods of the present invention can be taken from a variety of tissues, such as tumor tissue, blood and plasma, lymph, cerebrospinal fluid surrounding the brain and spinal cord, and synovial fluid surrounding the joint. In one embodiment, the sample is a blood sample. Blood samples are about 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0, 1.5, 2 0.0, 2.5, 3.0, 3.5, 4.0, 4.5, or 5.0 mL. The sample can be a tumor biopsy. The biopsy can be derived from, for example, the brain, liver, lung, heart, large intestine, kidney, or bone marrow. Any biopsy method used by those skilled in the art can be used to isolate a sample from a subject. For example, the biopsy can be a direct biopsy performed by a general anesthesiologist. The biopsy can also be a percutaneous biopsy performed with a smaller incision than that used for direct biopsy. The biopsy can also be a core biopsy or an incision biopsy where the tissue has been partially excised. The biopsy can also be an excision biopsy that attempts a complete excision of the lesion site. The biopsy can also be a fine needle aspiration biopsy where a tissue or body fluid sample is collected with a needle.

  In some embodiments, the clonotype profile of the method of the invention is generated from tumor or peripheral blood in the case of a diagnostic sample, or from peripheral blood in the case of a sample for monitoring residual disease. The One or more clonotypes associated with a disease such as a lymphoproliferative disease or a myeloproliferative disease are diagnosed from a diagnostic sample. Usually, one or more clonotypes associated with lymphoproliferative or myeloproliferative disorders are present as the most frequent in the clonotype profile. In some cases, there may be a single correlated clonotype, and in other cases there may be multiple clonotypes correlated with lymphoproliferative or myeloproliferative diseases. Tumor samples can be taken from any tissue affected by such diseases, including lymphatic vessels or other tissues outside the lymphatic system. As described above, a clonotype profile for monitoring residual lesions can also be generated from nucleic acid samples extracted from peripheral blood. The nucleic acid of the sample may be derived from B cells of the cell-containing fraction of peripheral blood, or may be derived from a cell-free nucleus of peripheral blood, such as plasma or serum. In one embodiment, the peripheral blood sample contains at least 1,000 B cells; more typically, such a sample contains at least 10,000 B cells, and more typically Contain at least 100,000 B cells. In another embodiment, the sample contains a large number of B cells in the range of 1000 to 1,000,000. In some embodiments, the number of cells in the sample sets the limit of measurement sensitivity. That is, by using a large amount of peripheral blood sample, the detection sensitivity of the residual lesion becomes higher. For example, in the case of a sample containing 1,000 B cells, the low frequency detection limit for clonotypes is independent of the number of sequencing reads obtained when analyzing the DNA of such cells by sequencing. 1/1000 or 0.001. The nucleic acid of the sample may be derived from T cells in a cell-containing fraction of peripheral blood, or may be derived from a cell-free nucleus of peripheral blood, such as plasma or serum. In one embodiment, the peripheral blood sample contains at least 1,000 T cells; more typically, such a sample contains at least 10,000 T cells, and more typically Contain at least 100,000 T cells. In another embodiment, the sample contains a large number of T cells in the range of 1000 to 1,000,000. In some embodiments, the number of cells in the sample sets the limit of measurement sensitivity. That is, by using a large amount of peripheral blood sample, the detection sensitivity of the residual lesion becomes higher. For example, in the case of a sample containing 1,000 T cells, the low frequency detection limit for clonotypes is independent of the number of sequencing reads obtained when analyzing the DNA of such cells by sequencing. 1/1000 or 0.001.

Samples used in the present invention may include DNA (eg, genomic DNA) or RNA (eg, messenger RNA). The nucleic acid may be cell-free DNA or RNA, for example, extracted from the circulatory system [Vlassov et al, CUIT. Mol. Med. 10: 142-165 (2010); Swarup et al, FEBS Lett. , 581: 795-799 (2007)]. In the methods provided by the present invention, the amount of RNA or DNA derived from a subject and that can be analyzed is, for example, in some applications as small as one cell (eg, other cell sorting criteria). , for example, the calibration test by morphological criteria), as well as an amount of ¹⁰⁶ cells more than this is the range of 6pg~60ug in terms of the amount of DNA in the range of 1pg~10ug in terms of the amount of RNA is there. In some embodiments, the nucleic acid sample is a 6 pg to 60 ug DNA sample. In other embodiments, the nucleic acid sample is a DNA sample from 100 μL to 10 mL of peripheral blood; in other embodiments, the nucleic acid sample is a DNA sample from 100 μL to 10 mL of peripheral blood cell-free nuclei.

In some embodiments, a sample of lymphocytes or cell-free nucleic acid presents substantially all B cells or T cells with different clonotypes and generates a clonotype “repertoire” based thereon. Therefore, it is large enough. In one embodiment, in order to substantially represent all the different chronotypes, a sample is taken with a 99% probability that each population chronotype present at a frequency of 0.001% or higher is included. . In another embodiment, samples are taken to have a 99% probability of containing each group's clonotypes present at a frequency of 0.0001% or greater. Also, in another embodiment, samples are taken so that each population that is present at a frequency of 0.00001% or more contains a 99% probability of clonotypes. In one embodiment, the B cell or T cell sample contains at least 5 × 10 ⁵ cells, and in another embodiment, such sample contains at least 1 × 10 ⁶ cells.

  Nucleic acid samples can be obtained from peripheral blood using conventional methods, for example, edited by Innis et al., PCR protocol (Academic Press, 1990). For example, white blood cells can be separated from a blood sample using conventional methods, for example, RosetteSep kit (Stem Cell Technologies, Vancouver, Canada). The blood sample can be in a dose range of 100 [mu] L to 10 mL; in one aspect, the blood sample dose is in the range of 100 [mu] L to 2 mL. The DNA and / or RNA can then be extracted from such blood samples using conventional methods such as, for example, the DNeasy Blood & Tissue kit (Qiagen, Valencia, Calif.) For use in the method of the present invention. . Optionally, using conventional methods such as, for example, fluorescence activated cell sorting (FACS) (Becton Dickinson, San Jose, Calif.), Magnetic cell separator (MACS) (Miltenyi Biotec, Auburn, Calif.), White blood cells For example, a subset of lymphocytes can be further isolated. For example, memory B cells can be isolated based on the surface markers CD19 and CD27.

  Cell-free DNA can be obtained using conventional methods such as Lo et al, US Pat. No. 6,258,540; Huang et al, Methods Mol. Biol. , 444: 203-208 (2008), and the like. As a non-limiting example, peripheral blood is collected in a blood collection tube containing EDTA, and then fractionated into a plasma component, a white blood cell component, and a red blood cell component by centrifugation. The DNA of the cell-free plasma fraction (eg, 0.5-2.0 mL) can be extracted using a kit such as the QIAamp DNA Blood Mini kit (Qiagen, Valencia, CA) according to the manufacturer's instructions. .

In one aspect, the lymphocyte sample to generate the clonotype profile is sufficiently large to allow presentation of substantially all T or B cells having different clonotypes. In one embodiment, samples are taken to have a 99% probability that each population clonotype present at a frequency of 0.001% or greater is included. In another embodiment, samples are taken to have a 99% probability of containing each group's clonotypes present at a frequency of 0.0001% or greater. In another embodiment, the sample is taken to contain a 99% probability of each population chronotype present at a frequency of 0.00001% or greater. In other embodiments, samples are taken to have a 95% probability of containing each group's clonotype present at a frequency of 0.001% or greater. In another embodiment, samples are taken to have a 95% probability that each population chronotype present at a frequency greater than or equal to 0.0001%. In another embodiment, the sample is taken to contain a 95% probability of each population chronotype present at a frequency of 0.00001% or greater. In yet another embodiment, the B cell or T cell sample contains at least 5 × 10 ⁵ cells, and in another embodiment, such sample contains at least 1 × 10 ⁶ cells.

  If the source of the material from which the sample is collected is scarce, for example in the case of clinical specimens, whole genome amplification (WGA), multiple displacement amplification (MDA); or similar techniques, such as Hawkins et al, Curr. Opin. Biotech. 13: 65-67 (2002); Dean et al, Genome Research, 11: 1095-1099 (2001); Wang et al, Nucleic Acids Research, 32: e76 (2004); and Hosono et al, Genome Research, 13 : The DNA derived from the material may be amplified by a technique that does not apply a bias, such as 954-964 (2003).

  The blood sample is of a specific subject and can be obtained using conventional methods such as, for example, edited by Innis et al., PCR Protocols (Academic Press, 1990). For example, white blood cells can be separated from a blood sample using conventional methods, for example, RosetteSep kit (Stem Cell Technologies, Vancouver, Canada). The blood sample can be in a dose range of 100 [mu] L to 10 mL; in one aspect, the blood sample dose is in the range of 100 [mu] L to 2 mL. The DNA and / or RNA can then be extracted from such blood samples using conventional methods such as, for example, the DNeasy Blood & Tissue kit (Qiagen, Valencia, Calif.) For use in the method of the present invention. . Optionally, using conventional methods such as, for example, fluorescence activated cell sorting (FACS) (Becton Dickinson, San Jose, Calif.), Magnetic cell separator (MACS) (Miltenyi Biotec, Auburn, Calif.), White blood cells For example, a subset of lymphocytes can be further isolated.

  Reconstitution has been identified in the DNA of each individual's acquired immune cells and their associated RNA transcripts, and in either case RNA or DNA is sequenced by the methods provided in the present invention. be able to. Reconstituted sequences from T cells or B cells that encode T cell receptor chains or immunoglobulin molecule chains, or portions thereof, are referred to as clonotypes. The DNA or RNA may correspond to a sequence derived from a T cell receptor (TCR) gene or an immunoglobulin (Ig) gene encoding the antibody. For example, DNA and RNA may correspond to sequences that can encode the α, β, γ, or δ chain of the TCR. In most T cells, the TCR is a heterodimer composed of α and β chains. The TCR α chain is generated by VJ rearrangement and the β chain receptor is generated by V (D) J rearrangement. Regarding the TCR β chain, there are 48 V fragments, 2 D fragments, and 13 J fragments in humans. It is also possible to delete some bases and add other bases (called N and P nucleotides) at each of the two junctions. In a small number of T cells, the TCR is composed of a γ chain and a δ chain. The TCRγ chain is generated by VJ reconstruction, and the TCRδ chain is generated by V (D) J reconstruction (Kenneth Murphy, Paul Travers, and Mark Walport, Janeway's Immunology 7th edition, Garland Science, 200 Which is incorporated herein by reference in its entirety).

  DNA and RNA to be analyzed in the method of the present invention include heavy chain immunoglobulin (IgH) having a constant region (α, δ, ε, γ, or μ) or light chain immunoglobulin having a constant region λ or K (IgK or IgL) may correspond to the sequence encoding. Each antibody has two identical light chains and two identical heavy chains. Each chain is composed of a constant (C) region and a variable region. For the heavy chain, the variable region is composed of a variable (V) segment, a diversity (D) segment, and a linked (J) segment. There are separate sequences in the genome that encode each of these segments. The occurrence of specific VDJ reconstitution events during B cell differentiation causes the cells to produce specific heavy chains. Diversity occurs in the light chain in a similar manner, except that there is no D region and only VJ rearrangement. In many cases, somatic mutations occur in the vicinity of the reconstitution site, which adds or deletes some nucleotides, further increasing the diversity of heavy and light chains produced by B cells. The diversity of antibodies produced by B cells can also be caused by differences in heavy and light chains. The variable region of the heavy and light chains is responsible for forming the antigen recognition (or binding) region or site. In addition to this diversity, somatic hypermutation processes can occur after specific responses that occur against several epitopes.

  As described above, in accordance with the present invention, primers can also be selected to generate amplicons containing reconstituted nucleic acids derived from lymphocytes or cell-free nucleic acid moieties derived from tissues such as blood. In the present specification, such a portion may also be referred to as a “region reconstructed by somatic mutation”. The region reconstituted by somatic mutation may contain nucleic acids derived from differentiating lymphocytes or fully differentiated lymphocytes, wherein the differentiating lymphocytes are molecules (eg, complete V (D ) Refers to a cell in which reconstitution of the immune gene forming the J region) has not been completed. For example, reconstituted regions where somatic mutation has not been completed are incomplete IgH molecules (such as molecules containing only the DJ region), incomplete TCRδ molecules (such as molecules containing only the DJ region) And inactive IgK (eg, including the Kde-V region).

Amplification of nucleic acid groups In some embodiments, the primer sequences of the first and second primer sets can also be selected according to conventional multiplex polymerase chain reaction (PCR). For example, guidance on primer selection and performance of multiplex PCR of nucleic acids encoding various immunoreceptor chains can be found in the following references, which are incorporated by reference: Faham And Willis, US Pat. Nos. 8,236,503 and 8,628,927; Morley, US Pat. No. 5,296,351; Gorski, US Pat. No. 5,837,447; Dau, US Pat. Van Dongen et al, US Patent Publication No. 2006/0234234; European Patent Publication No. 1544308B1; and Van Dongen et al, Leukemia, 17: 2257-2317 (2003). Guidance for multiplex PCR can also be found in references such as Henegariu et al, BioTechniques, 23: 504-511 (1997). In some embodiments, the primers are selected such that the sequence frequency that is amplified in the final product is substantially the same as the sequence frequency in the starting reaction mixture. Such primer selection can also include selection of primer length, primer binding site, and primer concentration. As described above, the multiplex level can vary widely depending on the method selected for creating the sequence read and the sequence tag added.

  In some embodiments, amplifying the target nucleic acid comprises, for example, annealing a set of primers (eg, a first set of “upstream” or “downstream” primers), extending the primer, and extending the primer. It involves linear amplification of the target nucleic acid, such as amplifying a large number of extended strands as a linear function of the number of cycles by repeating the cycle of melting the nucleic acid strand from the template. In other words, the amplification step involves replicating the target polynucleotide (ie, at least one target polynucleotide strand) by repeatedly extending a set of primers. In some embodiments, after such a single or repeated extension in one direction, removing the non-extended primer and another set of primers (eg, a second set of “downstream” or “ The step of extending the “reverse” primer) in a different direction once or repeatedly can also be performed.

  The number of primers in the first set of primers and the second set of primers can vary depending on the number and type of immunoreceptor chain nucleic acids amplified in the assay. In some embodiments, consensus primers may be used for the various strands. In other embodiments, specific primers can be designed for each different target polynucleotide to be amplified. Typically, both the first and second sets of primers each comprise a plurality of primers. In some embodiments, the plurality of primers in the first set or second set of primers is at least 50 primers; in other embodiments, the plurality of primers in the first set or second set of primers is In other embodiments, the plurality of primers in the first set or second set of primers is at least 150 primers; in other embodiments, the first set or second set of primers The plurality of primers in is at least 200 primers; in other embodiments, the plurality of primers in the first set or second set of primers is at least 250 primers. The number of primers in the first set of primers may be the same as or different from the number of primers in the second set of primers.

  In some embodiments, the first set of primers and the second set of primers are selected such that the clonotype is at least 30 bases in length; in other embodiments, the first set of primers and the second set of primers The set of primers is selected such that the clonotype is in the range of nucleotides 30-500 bases; in other embodiments, the first set of primers and the second set of primers have a clonotype of nucleotides 30-400. In other embodiments, the first set of primers and the second set of primers are selected so that the clonotype has a length in the range of 30-300 bases of nucleotides; In other embodiments, the first set of primers and the second set of primers have a It is selected to be the length of the range of fault 30 to 200 bases.

Examples of PCR amplification protocols can be found in van Dongen et al, Leukemia, 17: 2257-2317 (2003) or van Dongen et al, US Patent Publication 2006/0234234, which are incorporated by reference. . Briefly, examples of protocols include the following: Reaction buffer: ABI Buffer II or ABI Gold Buffer (Life Technologies, San Diego, Calif.); Final dose 50 μL; sample DNA 100 ng; each primer 10 pmol (as follows: Adjust to regulate amplification); dNTP final concentration 200 μΜ; MgCl ₂ final concentration 1.5 mM (optimize depending on target sequence and polymerase); Taq polymerase (1-2 U / tube); Cycle conditions: 7 at 95 ° C. Heat denaturation for minutes; annealing at 60 ° C .; number of cycles: denaturation 30 seconds; annealing 30 seconds; extension 30 seconds. The polymerase that can be used for amplification in the method of the present invention may be a commercially available one, and examples thereof include Taq polymerase, AccuPrime polymerase, or Pfu. The polymerase used can be selected based on fidelity or efficiency preference.

  In the initial step, the functional quantity of the amplifiable material in the sample can be determined by using real-time PCR, picogreen staining, electrophoresis using a fluid at the nanoscale (for example, LabChip) or ultraviolet absorption measurement.

  In one aspect, the multiplex amplification of the invention is performed such that the relative amount of sequences in the initial group is substantially equal to that in the amplified group or amplicon. That is, multiplex amplification is performed so that amplification bias between member sequences included in the sample group is minimized. In one embodiment, such relative amounts are substantially equal if each relative amount in the amplicon falls within five times the value in the initial sample. In another embodiment, such relative amounts are substantially equal if each relative amount in the amplicon falls within twice the value in the initial sample. As disclosed in detail below, amplification bias in PCR can be detected and corrected using conventional methods, so that a set of PCR primers can be determined to provide unbiased amplification for any sample. You can choose for your repertoire.

  In some embodiments, amplification bias can be avoided by performing a two-stage amplification (eg, as described in Faham and Willis, cited above), where the first or primary stage is Using a primer with a tail that is non-complementary to the target sequence, the amplification cycle is performed a few times (eg, 2-5, or 2-10, or 2-15 cycles). The tail includes a primer binding site added to the sequence amplicon of the primary amplicon, and such binding site is used for the second stage amplification using only a single forward primer and a single reverse primer. This eliminates the main cause of the amplification bias. Prior to initiating the second stage amplification, the primer not extended in the first stage is removed from the reaction mixture or inactivated. In some embodiments, the primary PCR has a sufficiently low number of cycles (eg, 2-10 cycles) to minimize differences that occur in amplification due to different primers. Next, secondary amplification is performed with a pair of primers to eliminate the occurrence of differential amplification. In some embodiments, several percent or several parts of the reaction dose, eg, 1% of the primary PCR reaction dose, is used directly in the secondary PCR reaction mixture. In some embodiments, a total of at least 35 cycles is allocated to the first stage amplification and the second stage amplification.

  In some embodiments, internal standards can be combined and amplified in the same reaction as the reconstituted nucleic acid sample. The internal standard is a nucleic acid of known sequence and known concentration. For example, the internal standard can be a duplicate of a natural nucleic acid encoding a portion of an immunoreceptor chain, or it can be a synthetic nucleic acid. In some embodiments, the length and base composition of the internal standard is selected to reproduce the specific immunoreceptor chain to be amplified. By monitoring the change in the relative concentration of the internal standard after amplification, the amplification bias can be detected, and amplification conditions without bias can be determined. For example, primer length, position, and concentration can be altered to minimize bias of the amplification product. In some embodiments, multiple internal standards are used in the reaction; in some embodiments, 2-50 different internal standards are used in the reaction; in other embodiments, 2-25 different standards are used. An internal standard is used for the reaction; in some embodiments, 2-10 different internal standards are used for the reaction. In some embodiments, the amplification bias is determined by measuring the relative frequency of sequences of different target nucleotides (eg, all or excerpted chronotypes or internal standards) in the amplification product. In other embodiments, the presence, absence, or level of amplification bias is determined by quantitative real-time PCR of selected nucleic acids, such as two or more internal standards. An internal standard can also be used to quantify the number of different clonotypes contained in the original sample. Such molecular quantification methods are well known and include, for example, Brenner et al, US Pat. No. 7,537,897. This document is incorporated herein by reference.

Generation of Sequence Reads Any high throughput nucleic acid sequencing method can be used in the methods of the present invention. Preferably, such methods are capable of generating large amounts of sequence data in a cost-effective manner, so that at least 1000 chronotypes can be determined, preferably at least 10,000. ~ 1,000,000 chronotypes can be produced. DNA sequencing methods include labeled dideoxy methods using labeled ends or primers and slab gel and capillary gel separation (Sanger method), synthetic sequencing using reversibly end-labeled nucleotides, pyrosequences, 454 Sequencing, allele-specific hybridization to a library of labeled oligonucleotide probes, allele-specific hybridization to a labeled clone library and subsequent sequencing using ligation, of labeled nucleotides incorporated during the polymerization process Real-time monitoring, polony sequencing, and SOLiD sequencing. In recent years, it has been demonstrated to sequence separated molecules by sequential or single extension reactions using polymerases or ligases, as well as single or continuous differential hybridization using probe libraries. These reactions have been performed simultaneously on a large number of clonal sequences so far, and those currently demonstrated in parallel with 100 × 10 ⁶ sequences in current commercial applications. Thus, these sequencing approaches can be used in studying the T cell receptor (TCR) and / or the B cell receptor (BCR) repertoire. One aspect of the invention utilizes a high-throughput sequencing method that includes the step of spatially separating individual molecules on a solid phase surface that is simultaneously sequenced. Such solid phase surfaces are non-porous surfaces [Solexa sequencing, eg Bentley et al, Nature, 456: 53-59 (2008) or Complete Genomics sequencing, eg Dranac et al, Science, 327: 78. 81 (2010)], an array of wells that may contain a template with beads or particles attached [such as by 454 companies, eg, Margulies et al, Nature, 437: 376-380 (2005) or Ion Torrent sequencing. , US Patent Publication No. 2010/0137143 or 2010/0304982], microfabricated membranes [SMRT sequencing, eg, Eid et al, Science, 323: 133-1 38 (2009)], or bead arrays (SOLiD sequencing or polony sequencing, eg, Kim et al, Science, 316: 1481-1414 (2007)). Amplifying the isolated molecules before or after spatially separating them from each other on the solid surface, which may include emulsion-based amplification, such as emulsion PCR, or rolling circle type amplification. You can also.

After sequencing by synthesis using a reversible terminator, such as Solexa sequencing, in which individual template molecules are spatially separated on the solid surface, the molecules are simultaneously amplified by bridge PCR to produce separate clones Bentley et al (supra) and manufacturer's instructions (eg, TruSeq ™ sample preparation kit and data sheet [Illumina, San Diego, Calif., 2010]). And the following references: US Pat. No. 6,090,592; US Pat. No. 6,300,070; US Pat. No. 7,115,400; and European Patent No. EP 0 972 081 B1 (these references are incorporated by reference) The approach using the description in . Interest In one embodiment, each molecule is located and amplified to a solid surface, a density of at least ¹⁰ 5 / cm ^2; or at least 5 × ¹⁰ 5 / cm ² density, or at least ¹⁰ 6 / cm ² Solexa-based sequencing generates two sequence reads from the same target sequence (or template) in the cluster and one sequence read from the opposite end of each target sequence In some embodiments, such sequence read pairs can also be combined and treated as a single sequence read in a series of analyses, or such pairs can be separated separately, although they Can be processed taking into account that they are derived from the same cluster, optionally sequence reads from the same template Pairs are referred to as “mate pairs,” and the sequencing process from both ends of the template is referred to as “bidirectional” sequencing.In some embodiments, synthesis using reversibly end-labeled nucleotides. The steps of sequencing by: a single sequence read for each cluster or clone group of templates and multiple sequence reads for each cluster or clone group of templates (including but not limited to mate pairs) In yet a further embodiment, when generating multiple sequence reads for each cluster or clone group of a template, such multiple sequence reads can be combined for subsequent analysis such as a binding step. It can also form a single sequence read that is useful to use Kill.

  In one aspect, a clonotype profile based on the sequence of a sample derived from an individual comprises the following steps: (a) obtaining a nucleic acid sample derived from the individual's T cells and / or B cells; (b) such a nucleic acid. Spatial separation of individual molecules from the sample (individual molecules include at least one template generated from nucleic acids in the sample, which is a region reconstructed by somatic mutation or those Each individual molecule is capable of producing at least one sequence read); (c) sequencing the spatially separated individual molecules; and (d) a nucleic acid Assessing the abundance of nucleic acid molecules in each sequence from the sample to generate a clonotype profile. In one embodiment, the region reconstructed by somatic mutation comprises a V region and a J region, respectively. In another embodiment, the sequencing step includes generating a plurality of sequence reads for each determined clonotype. In yet another embodiment, the sequencing step includes combining information or data from multiple sequence reads to generate each chronotype. In some embodiments, such a combining step involves combining sequence reads as described in Faham and Willis, US Pat. No. 8,628,927, the teachings of which are hereby incorporated by reference. Or by using a sequence tag as described in Faham et al, US 2013/0236895 A1, the teachings of which are incorporated herein by reference. In another embodiment, sequencing comprises bi-directionally sequencing each of the spatially separated individual molecules to generate at least one forward sequence read and at least one reverse sequence read. Including doing.

  Further, in the latter embodiment, at least one forward sequence read and at least one reverse sequence read have overlapping regions, and the bases of such overlapping regions are complementary in the reverse direction between such sequence reads. It is decided based on the relationship. In yet another embodiment, each of the regions reconstructed by somatic mutation comprises a V region and a J region, and the sequencing step starts from a portion in the J region and the direction of the associated V region Further comprising determining the sequence of each of the individual nucleic acid molecules based on the one or more forward sequence reads and at least one reverse sequence read extending to In another embodiment, the individual molecules are complete IgH molecules, incomplete IgH molecules, complete IgK molecules (complete), inactivated IgK molecules, TCRβ molecules, TCRγ molecules, complete TCRδ molecules, and A nucleic acid selected from the group consisting of incomplete TCRδ molecules. In another embodiment, sequencing includes generating a sequence read with a monotonically decreasing quality score. In another embodiment, the above method comprises the following steps: (a) obtaining a nucleic acid sample derived from an individual's T cells and / or B cells; (b) individual molecules derived from such a nucleic acid sample. (Individual molecules each include a template nested set generated from the nucleic acid in the sample and each region or part thereof reconstructed by somatic mutation, and each nested set Can produce multiple sequence reads, each extending in the same direction and starting from different parts on the nucleic acid from which the nested set was generated); (c) sequencing this spatially separated individual molecule And (d) evaluating the abundance of nucleic acid molecules of each sequence from the nucleic acid sample to generate a clonotype profile.In another embodiment, the sequencing step includes generating a plurality of sequence reads for each of the nested sets. In another embodiment, each region reconstructed by somatic mutation comprises a V region and a J region, each of the plurality of sequence reads starting from a different portion of the V region and associated J Stretch in the direction of the region.

  In one aspect, for each sample from an individual, the sequencing method used in the method of the invention produces at least 1000 clonotype sequences per run; in another aspect, such an approach comprises: Producing at least 10,000 clonotype sequences per run; in another aspect, such an approach produces at least 100,000 clonotype sequences per run; in another aspect, such an approach Produces at least 500,000 clonotype sequences per run; as well as in other embodiments such an approach produces at least 1,000,000 clonotype sequences per run. In yet another aspect, such an approach produces 100,000 to 1,000,000 chronotype arrays per run of individual samples. In each of the foregoing, each chronotype per run is determined from at least 10 sequence reads.

  About 30 bp, about 40 bp, about 50 bp, about 60 bp, about 70 bp, about 80 bp, about 90 bp, about 100 bp, about 110, about 120 bp, about 150 bp per read, depending on the sequencing method used in the provided method of the present invention About 200 bp, about 250 bp, about 300 bp, about 350 bp, about 400 bp, about 450 bp, about 500 bp, about 550 bp, or about 600 bp.

Determination of clonotype from sequence data In some embodiments of the invention, a sequence tag is used to determine the clonotype, and in other embodiments, the step of combining the sequence tag in combination with a sequence read is used, Determine the chronotype. In embodiments where a single unique sequence tag is added to substantially all different target polynucleotides, the clonotype determination by the sequence tag is straightforward. In such an embodiment, the sequence reads are first grouped based on the sequence tag to determine the chronotype of the sample. Such grouping can also be performed by a conventional sequence alignment method. As a guideline for selecting an alignment method, those described in Batzoggu Briefings in Bioinformatics 6: 6-22 (2005), which are incorporated herein by reference, can be used. After constructing the sequence reads into groups according to unique sequence tags, the relevant clonotype sequences can be analyzed to determine the clonotype sequence from the sample. FIG. 4A illustrates an example of an alignment and method by determining the chronotype sequence (SEQ ID NO: 2) associated with a unique sequence tag. In this example, after aligning 11 sequence reads based on each sequence tag (4302), the nucleotides of each portion of the clonotype portion (4304) of each sequence read are 1, 2, 3, 4 ,. . . Shown as n and compared. For example, the nucleotides of moiety 6 (4306) are t, t, g, t, t, t, t, t, t, c, t; that is, nine base calls are t and one is “g (4308), and one is “c” (4310) (SEQ ID NO: 3 and SEQ ID NO: 4). In one embodiment, the correct base designation at each position of the clonotype sequence depends on how the primary base is identified. In the example of part 6 (4306), the base designation is “t” since the nucleotide of this part is t in most of the sequence reads. In other embodiments, an accurate base call can be determined for a clonotype sequence, taking into account other factors such as the sequence read base call quality score or identification of adjacent bases. Once the chronotype is determined as described above, a chronotype profile can be constructed that includes the abundance or frequency of each different chronotype in the sample.

  In some embodiments, one or more extension steps are utilized utilizing primers containing a sequence tag to increase the target polynucleotide fraction in the sample that is labeled with the sequence tag prior to amplification. A process can also be carried out. In such embodiments, the target polynucleotide and / or a copy thereof is labeled with a plurality of different sequence tags by one or more extension steps in the presence of a primer containing the sequence tag. “Multiple” refers to the number of extension steps performed in the presence of a primer containing a sequence tag, the efficiency of the amplification reaction, and whether the forward or reverse primer has a sequence tag or both are sequenced It depends on whether you have a tag. In some such embodiments, “plurality” is in the range of 2-15, or in the range of 2-10, or in the range of 2-5. In some such embodiments, after amplification, each target polynucleotide copy of the sample may be classified into a plurality of groups or subsets, wherein members of each group or subset are the same sequence tag. Members of each different group or subset are labeled with different sequence tags; that is, members of the same group have the same sequence tag and members of different groups have different sequence tags. In other words, in some embodiments, after amplification, each copy of the target polynucleotide from the sample will be labeled with one of two different sequence tags; or in other embodiments, Each copy of the target polynucleotide from the sample will be labeled with one of three different sequence tags; or in other embodiments, each copy of the target polynucleotide from the sample is different. It will be labeled with one of the four sequence tags, and this will be repeated sequentially thereafter. In these embodiments, clonotypes are derived from the same parent sequence based on the likelihood that after aligning the sequence tag combination, the common origin is true as a function of error rate and relative frequency. It can also be determined by the step of combining to process the sequence reads within the group. FIG. 4B illustrates sequence reads from such an embodiment. In one approach, sequence reads are first grouped by a common sequence tag (4402), which in the case of this figure results in the formation of three groups (4420), (4422), and (4424). In some embodiments, within each group, the sequence (4404) is analyzed to determine the group's consensus sequence; for example, as described above, in each nucleotide portion, the base is the major base, or the most It can be referred to as a high frequency base. The consensus sequence groups can then be combined together to determine the chronotype.

  In some embodiments, the above aspects of the invention can also be implemented in a method for profiling virtually any group of nucleic acids in a sample. Such a method comprises the steps of: (a) obtaining a sample comprising a group of nucleic acids; (b) adding a sequence tag to the group of nucleic acids to form a tag-nucleic acid complex, wherein at least one Nucleic acid groups or copies thereof are appended with different sequence tags); (c) amplifying the tag-nucleic acid complex; (d) sequencing the tag-nucleic acid complex to provide an error rate nucleic acid sequence And (e) aligning sequence reads having similar tag sequences to form a group of sequence reads having the same sequence tag; (f) creating a group of sequence reads. Binding to determine a nucleic acid sequence, wherein whenever the likelihood of a group of sequence reads is clearly at least 25%, the group of sequence reads is bound to a different sequence; Determining the sequence profile of the group by determining the level of g) sequences including,. When utilized for profiling of reconstituted nucleic acid groups, such a method comprises the following steps: (a) obtaining a sample containing T cells and / or B cells and / or cell-free DNA from an individual; (B) adding a sequence tag to the reconstituted nucleic acid molecule of the sample-derived T cell receptor gene or immunoglobulin gene to produce a tag-nucleic acid complex (wherein at least one reconstitution from the sample) (C) amplifying the tag-nucleic acid complex; (d) sequencing the tag-nucleic acid complex and having an error rate. Generating a sequence read comprising a tag sequence and a reconstructed nucleic acid sequence; (e) aligning sequence reads having similar tag sequences and grouping sequence reads having the same sequence tag (F) combining the group of sequence reads to determine the clonotype (where the group of sequence reads is clearly at least 25% and the group of sequence reads is a different sequence) And (g) determining the clonotype profile of the sample by determining the level of the clonotype.

In the above and other embodiments disclosed herein, sequencing the tag-nucleic acid complex comprises sequencing a sample of the tag-nucleic acid complex derived from an amplicon. Typically, such samples maintain the relative frequency of the target polynucleotide in the original sample (ie, tissue sample or blood sample, etc.) also in the tag-nucleic acid complex sample derived from the amplification reaction product. A representative sample. In some embodiments, a reconstituted nucleic acid group encoding an immunoreceptor molecule is analyzed, and the tag-nucleic acid complex sample comprises at least 10 ⁴ tag-nucleic acid complexes; In form, such a sample comprises at least 10 ⁵ tag-nucleic acid complexes; in other embodiments, such a sample comprises at least 10 ⁶ tag-nucleic acid complexes; Then, such a sample contains at least 10 ⁷ tag-nucleic acid complexes.

Binding Sequence Leads In embodiments where multiple sequence tags are added to the original reconstructed nucleic acid or a copy thereof, binding the sequence reads (or group consensus sequence reads) to determine clonotypes. May be implemented. If the sequencing method contains no errors, it is obvious to reduce the set of sequence reads for a given sample to a different set of chronotypes and to record the number of reads for each chronotype. However, if there are sequencing errors, each real clonotype will be buried in a “population” of sequence reads that contain various numbers of errors in the sequence. A “population” of sequencing errors increases the distance from the clonotype of the sequence space, reducing density. Various algorithms are available for converting sequence reads into chronotypes. In one aspect, binding of sequence reads (ie, combining clonotype candidates that have been confirmed to have one or more sequence errors) is at least three factors: the number of sequences obtained for each clonotype to be compared Depending on the number of different bases; and the sequencing quality score of the mismatched part. In some embodiments, a likelihood ratio may be constructed and evaluated based on the expected error rate and the binomial distribution of errors. For example, two chronotypes, one with 150 reads and one with two leads, with one difference in the region with low sequencing quality, can be generated by sequencing errors. There is a tendency to be combined as having nature. On the other hand, two chronotypes, one with 100 differences and one with 50 leads, that contain two differences, are considered unlikely that these chronotypes were generated by sequencing errors. Is not combined. In some embodiments, the following algorithm can also be used in determining clonotypes derived from sequence reads. Some of these concepts are illustrated in FIG. 5A. In some embodiments of the binding step, the sequence reads are first converted to candidate clonotypes. Such a conversion depends on the sequencing platform utilized. For platforms that generate long sequence reads with high Q scores, sequence reads or portions thereof can also be directly employed as clonotype candidates. For platforms that generate short sequence reads with a low Q score, several alignment and construction steps may be required to convert the set of related sequence reads into candidate clonotypes. For example, in some embodiments, as for the Solexa-based platform, as described above, chronotype candidates are generated based on a collection of leads paired from, for example, 10 or more clusters.

  As illustrated in FIG. 5A, the frequency of clonotype candidates can be plotted in sequence space. In FIG. 5A, such a space is corrected to one dimension (horizontal axis) for convenience of illustration. The vertical axis gives the frequency of each chronotype candidate, log (read count), or the magnitude of a similar measurement. In the figure, chronotype candidates are represented by various symbols (530). According to one embodiment of the present invention, two candidate chronotypes are assigned their respective frequency or read count (as described above), the number of different bases between them (the larger the different number, As well as the quality score of bases at different positions in each sequence (the higher the quality score, the lower the possibility of binding). Chronotype candidates can be considered in order of their respective frequencies. FIG. 5A shows chronotype candidate 1 (532), chronotype candidate 7 (534), and chronotype candidate 11 (536) as the three most frequent candidates. In relation to each such chronotype candidate, there are other chronotype candidates that are close in sequence but inferior in frequency, eg, for (i) chronotype candidate 1 (532), chronotype candidate 2 ( 538) and Chronotype Candidates 3, 4, 5, and 6 contained within Cone (540); for Chronotype Candidate 7 (534), contained within Cone (542), There are chronotype candidates 8, 9 and 10; (iii) For chronotype candidate 11, there is chronotype candidate 12 included inside cone (544). This cone shows the boundary of likelihood, and the chronotype candidates contained within the boundary will be combined with the more frequent chronotype candidates 1, 7 or 11. Such likelihood boundaries depend on the frequency of neighboring clonotype candidates (3, 4, 5 and 6 for 1; 8, 9 and 10 for 7; and 12 for 11) and in sequence space, respectively. Depending on the distance from each of the more frequent chronotype candidates. Chronotype candidate 2 (538) is outside of cone (540); therefore, it is not combined with chronotype candidate 1 (532). The higher the frequency of candidate clones, the more likely it is a truly different chronotype than the less frequent one, and the less frequent (several) differences than the more frequent (several) differences Since the possibility of an error increases, the (boundary) likelihood boundary is again shown as a cone.

  The sequence read population that embeds each clonotype candidate can be modeled using a simple model for binomial distribution and the likelihood of single base error. The latter error model can be inferred from mapping of V and J fragments, or can be inferred from self-alignment and convergence from the chronotype itself that finds the algorithm. The model is a null model where only X is a true clonotype in this region of the sequence space, and given a “population” sequence Y with a read count C2 and an E error (for sequence X): It is constructed based on the probability of being part of a true chronotype sequence X having a complete read count C1. According to the parameters C1, C2 and E, it is determined whether or not the sequence Y is bound to the clonotype X. For any given C1 and E, the maximum value C2 is pre-calculated to determine the binding of array Y. Under the null hypothesis that Y is included in chronotype X, after putting together all possible sequences Y that have error E into adjacent sequences X, less than the P value that is likely to fail to join Y The maximum value of C2 is selected so that The P value is a value that controls the behavior of the algorithm and increases or decreases binding tolerance.

  In binding to clonotype X, if sequence Y is not bound to clonotype X because its read count exceeds threshold C2, sequence Y is a candidate for providing another clonotype (in FIG. 5A, clonotype X). Type candidate 2 (538), etc.). Similarly, in an algorithm that implements such a principle, any other array Y2, Y3, etc. that is “closer” to this array Y (which is considered to be independent of X) Certainly it will not be collected into X. This concept of “similarity” includes both error counts for Y and X and complete read counts for X and Y, ie, the same model as above for the error sequence population around clonotype X. Modeled in the way. In this way, if it happens to be “close” to one or more chronotypes, the “population” sequence may be appropriately attributed to their exact chronotype. Thus, according to FIG. 5A, if Chronotype Candidate 2 is considered to be truly different from Chronotype Candidate 1 (532), then by a unique routine or sub-algorithm, Chronotype Candidate 1 (532) and 2 ( 538) and a rule is provided that determines which (and which) candidates 4 and 5 between 1 and 2 should be combined.

  In one embodiment, the algorithm proceeds from top down, starting with sequence X with the highest read count. This arrangement provides the first chronotype. If their count falls below a pre-calculated threshold (see above), adjacent sequences bind to the first clonotype, or if they exceed the threshold, or another sequence that did not bind If it is “closer”, leave it as it is. After searching all adjacent sequences within the maximum error count, the process of binding the lead to chronotype X is completed. Combined leads and all leads are added to and excluded from the list of other leads available for chronotype generation. The next sequence is then moved to the highest read count. Combine adjacent reads to this chronotype as described above and continue this process until there are no more sequences with read counts above the threshold, e.g., all sequences with counts greater than 1 are used as chronotype seeds. continue.

  As noted above, another embodiment of the algorithm further includes testing to determine whether the candidate sequence Y is bound to an existing clonotype X taking into account the quality score of the associated sequence read. May be added. For sequence (s) Y, determine the average quality score (s) (averaged over all reads with sequence Y). Here, the arrays Y and X are different. If the average score exceeds a predetermined value, the difference is likely to indicate a truly different clonotype and should not be combined, and if the average score is below the predetermined value, the sequence Y is sequenced It should be coupled to X because it is most likely caused by an error.

  In order to successfully implement the above algorithm for combining clonotype candidates, all sequences with lower E errors than some input sequences X (ie, less than some sequence distance measurements) are all efficiently A method of identifying is required. This problem can be solved by using an array tree. As illustrated in FIG. 5D, the implementation of such a tree has some unique characteristics that the nodes of the tree are not limited to single letter notation of the DNA sequence of the candidate chronotype. Nodes can have any long array that allows more efficient use of computer memory.

  Place all leads for a given sample in an array tree. Each leaf node has a branch point to the associated lead. Sequences unique to clonotype candidates are recovered by scanning backwards from leaf nodes to root nodes. Place the first array in a simple tree with one root node and one leaf node containing the complete array of leads. Arrays are added side by side. If the tree already contains sequences, for each added sequence, a new branch is formed at the end of the common sequence between the lead and the existing tree, or a lead is added to the existing leaf node . Placing all the leads in the tree makes it easier to use the tree for the following purposes 1), 2): 1) Identify the highest read count: Sort leaf nodes based on the read count Can identify leaf nodes with the most leads (ie, sequences) and leaf nodes with continuously fewer leads; 2) identify adjacent leaves: for any sequence, less than X for this sequence All paths in the tree with errors can be searched. The path starts at the root, and this path travels along the tree and branches to another path. As you progress along the tree, list the latest error count for each path. A given path ends when the error count exceeds the maximum allowed for errors. In this way, the majority of the tree is rounded up as soon as possible. This method efficiently identifies all paths (ie, all leaves) within the error of X based on any given sequence.

The features of the above concept are illustrated in detail by the flowchart of FIG. 5B. A set of candidate clonotypes is obtained from sequence data obtained by sequencing reconstituted nucleic acids extracted from T cell or B cell samples. In one aspect, each clonotype candidate includes an NDN region and portions of a V region and a J region. These arrays are structured into a data structure (550). This data structure may be an array tree. Although not shown in FIG. 5B, in one embodiment, a sequence tree may be constructed for known V regions and known J regions when generating a set of clonotype candidates. Based on the sequence tree, the sequence reads that make up the clonotype candidates are mapped or aligned to these known sequences, and the most likely known V and J sequences for the clonotype candidates are efficiently You may decide to. Returning to FIG. 5B, once a chronotype candidate has been generated, data such as a sequence tree for use in a method for identifying a true chronotype from a chronotype candidate that includes experimental or measurement errors (such as sequencing errors). Build the structure. From the data structure, eg, sequence tree, select the chronotype candidate that has the highest frequency of occurrence among the current chronotype candidates (HFCC _k ) (552); in other words, HFCC _k is the cycle among the chronotype candidates. In k, the copy number is the largest, that is, the read count is the largest. Next, adjacent, less frequent chronotype candidates are specified (LFCCs) (554); that is, chronotype candidates within a distance _Dk are specified. In one aspect of the invention, this identification is performed using a sequence tree that allows efficient sequence comparison of relatively short (less than 300 bp) sequences.

In one embodiment, the comparison or sequence alignment is performed using dynamic programming, such as those disclosed in Gusfield (supra). In a further embodiment, such dynamic programming is bandwidth limited dynamic programming and does not consider sequences that differ beyond a predetermined distance from the selected HFCC, thus increasing computational speed. . Candidate HFCC _k and LFCC _j can be compared based on many different criteria or attributes. In one aspect, as described above, clonotype candidates are compared based on at least two criteria: (i) frequency or read count and (ii) sequence differences. In one aspect, as described above, based on at least three criteria: (i) frequency or read count, (ii) sequence difference, and (iii) quality score or evaluation of the base in which the difference occurs, Compare type candidates. In one embodiment, sequence differences include base substitutions; in another embodiment, sequence differences include base substitutions, deletions and insertions. The latter embodiment is particularly applicable when the sequence data is generated by a synthesis-time decoding method that does not use a terminator, such as a 454 sequencer and an Ion Torrent sequencer. Such a sequencing approach distinguishes homopolymer extensions of different sizes based on the signal amplitude; therefore, the larger the homopolymer size, the greater the difference in signal level due to the difference in one homopolymer nucleotide. Because of the significant reduction (ie, 2-mer is easily distinguishable from 3-mer, 8-mer is almost indistinguishable from 9-mer), such an approach is Prone to insertion and deletion errors in routines. In one aspect, a function such as P (HFCC _k , LFCC _j , D, Q) that depends on the amount of (i) to (iii) shown in the decision box (558) (here, “join likelihood” HFCC and LFCC comparisons are performed using a "referred to as a degree function". Although such functions can take many different forms, in general, the P value varies with changes in (i), (ii) and (iii) as follows: Preferably, it increases monotonically by the HFCC frequency and the ratio of the HFCC frequency to the LFCC frequency, and the higher the ratio of the HFCC frequency to the LFCC frequency, the higher the likelihood that the LFCC is combined with the HFCC. Similarly, the P-value is preferably monotonically reduced with different degrees of HFCC and LFCC sequences, and the greater the difference between HFCC and LFCC (eg, the number relative to each other with the least number of substitutions, insertions or deletions). And the likelihood of coupling the LFCC to the HFCC is small. Finally, the P value decreases as the quality score at a position where the HFCC and LFCC sequences differ increases, preferably monotonically, and the higher the quality score, the smaller the likelihood of LFCC binding to HFCC.

When the HFCC and LFCC sequences differ at one or more positions, the quality scores at different positions can be combined in a variety of different ways. In one embodiment, whenever there are multiple such differences, the multiple quality scores are represented as an average value. This average value may be either an unweighted average or a weighted average. FIG. 5C shows an example function, P, that calculates the respective quality values (curves ae) for a given sequence difference. As shown in FIG. 5C, if the HFCC is on the order of about 200 read counts (570) and the quality score is determined by curve (a), any LFCC less than about 50 read counts (572) will be bound to the HFCC. The independent variable D of the function P is a measure of the distance between the arrays HFCC _k and LFCC _j , which can vary from cycle to cycle as analysis proceeds (“k” is appended with “k”). Meaning that the value of the constant can depend on the calculation cycle k). In one embodiment, D = D _k and this value is a function of the number of cycles. In another embodiment, D = D (HFCC frequency), and this value is a function of the HFCC frequency independent of the number of cycles. For example, as the frequency of HFCC decreases, the distance D of candidates to be compared decreases. In one embodiment, D is the Hamming distance between HFCC _k and LFCC _j ; however, other distance measurements may be used. In one embodiment, D _k is a non-increasing function of k; in another embodiment, D _k is a decreasing function of k. In some embodiments, it is useful to reduce the size of D as the number of cycles increases or the HFCC frequency decreases. Since the frequency of chronotype candidates decreases as the calculation progresses, most of these candidates are singletons, and thus sequence distance is the primary comparison (rather than frequency differences). As computation proceeds and D decreases, non-productive comparisons to far and less frequent chronotype candidates are reduced, thereby increasing computation speed. The function P can be a complex expression depending on the number of factors to be examined. FIG. 5C illustrates values calculated for P in one embodiment. P is related to the read count threshold when combining LFCC with a given read count for HFCCs with different quality scores, as described above. Curves “a” to “e” represent different quality score relationships (curve “a” corresponds to the highest quality score).

Returning to FIG. 5B, if P <P _k , LFCC _j is not combined with HFCC _k and another LFCC is selected (560). If P> P _k , LFCC _j is combined with HFCC _k (562), and another LFCC is selected (566) except when no LFCC to be evaluated remains (564). If there are no remaining LFCCs to evaluate (564), the current HFCC _k (including all combined LFCCs) is excluded from the data structure such as an array tree (568). Such exclusion is illustrated in the simple sequence tree (590) of FIGS. In the arrangement tree (590), the route (592) (indicated by a broken line) corresponds to the HFCC (596), and the LFCC (598) is coupled to this route. After combining, the path segment (592) in the shaded area (599) is excluded from the sequence tree (590), reducing the sequence tree (597) shown in FIG. 5E. After such removal utilizing this tree for a series of computations that identify neighboring LFCCs (554), if the stop rule (570) is satisfied, the chronotype determination is completed. In one embodiment, the stop rule (570) is whether the last non-singleton chronotype candidate has been processed (552). In another embodiment, the stop rule (570) is whether the frequency or read count of the selected HFCC is below the frequency or read count corresponding to the case of a single lymphocyte. In one aspect of the method of the invention, the amplification step results in each lymphocyte in the sample being represented by multiple copies of the same clonotype; thus, in one embodiment, the HFCC read, whatever If the count is below the read count corresponding to a single lymphocyte, the calculation is stopped. In some embodiments, such a read count number (or a copy of a clonotype candidate) is at least 10; in another embodiment, such a number is at least 20; , Such number is at least 30; in another embodiment, such number is at least 40. If the stop rule is not satisfied, the next HFCC is selected (572). The analysis process summarized in the flowchart of FIG. 5B can be implemented in any suitable programming language such as C, C ++, Java, C #, Fortran, or Pascal.

  According to one aspect of the invention, the above method for determining clonotype and / or clonotype profile comprises the steps of: (a) forming a recombinant immune molecule data structure from sequence reads obtained by high-throughput nucleic acid sequencing; (B) whenever the frequency of the inferior one is below the predetermined frequency value and the degree of their sequence difference is below the predetermined degree, the most frequent chronotype candidate and the more frequent Combining the inferior clonotype candidate to form a clonotype, (c) removing the combined clonotype candidate from the data structure, and (d) until a clonotype profile is formed (b ) And (c) are repeated. In one embodiment, the data structure is an array tree.

  According to another aspect of the invention, the above method of determining clonotype comprises: (a) providing a set of sequence reads from repertoires of recombinant immune molecules each having a V region, an NDN region, and a J region; Here, for each such molecule, at least one sequence read includes at least part of the NDN region of such molecule; (b) from a sequence read that includes at least part of the NDN region, Forming a sequence tree represented by the leaves [each leaf and the corresponding clonotype candidate has a frequency]; (c) any of the inferior frequencies below a predetermined frequency value and their sequences Whenever the degree of difference is less than a certain degree, the most frequent chronotype candidate and the less frequent chronotype candidate are combined. Forming a clonotype having the most frequent clonotype candidate sequences; (d) removing the combined leaves corresponding to the clonotype candidates from the sequence tree; and (e) less frequent. The process including the steps of repeating steps (c) and (d) until the maximum frequency of candidate chronotypes falls below a predetermined stop value. In one embodiment, the step of forming identifies all inferior chronotype candidates that have a different degree and a different sequence than the step of selecting the most frequent chronotype candidates. Forming a binding subset. Thus, in such an embodiment, the total number of LFCCs to be compared for the join process can be limited (only one within a predetermined different value is considered). Such a value is input from the process according to the application, and corresponds to, for example, the scale of the repatova and the calculation time used. As described above, the function used to determine whether to bind the LFCC to the HFCC can have a variety of forms. In one general aspect, in the combining step, this function may have the following attributes: HFCC frequency, LFCC frequency, differences between their sequences (such as Hamming distance, etc.) Depending on the quality score of one or more nucleotide positions (where HFCC and LFCC are different); the function is (i) the ratio of HFCC frequency to LFCC frequency Increases monotonically, (ii) decreases monotonically as the difference between HFCC and LFCC sequences increases, and (iii) decreases monotonically as the quality score at one or more nucleotide positions increases. That is, from the viewpoint of attribute (iii), if there is a high probability that HFCC and LFCC are different (for example, the reliability of the base call is very high), the possibility of combining them becomes low.

  In some embodiments, whenever such sequence reads have a apparent likelihood of at least 95 percent, binding such sequence reads to bind to different clonotypes (or target polynucleotides such as reconstituted nucleic acids). Choosing a likelihood function; in other embodiments, selecting a binding likelihood function to bind such sequence reads to different clonotypes whenever the likelihood of the sequence reads is clearly at least 99 percent In other embodiments, the binding likelihood function is chosen to bind such sequence reads to different clonotypes whenever the likelihood of the sequence reads is clearly at least 99.9 percent. As noted above, in some embodiments, the binding likelihood function may include the error rate of the chemical technique used for sequencing, the number of different nucleotides in the sequence reads being compared, as well as the relative sequence reads being compared. Depending on the frequency; in another embodiment, the binding likelihood function is the error rate of the chemical technique used for sequencing, the number of different nucleotides in the sequence reads being compared, the relative number of sequence reads being compared It depends on the frequency and the quality score of the different nucleotides. In the above, the predetermined frequency and the predetermined degree of difference are selected according to a specific application at the time of design. Factors that influence such selection include details of biological phenomena and implementation speed.

Monitoring Applications In one aspect, the present invention is directed to a method of monitoring minimal residual disease by assessing the presence, absence, and / or level of nucleic acids characteristic of or associated with a disease in a sample. To do. In some embodiments, such a nucleic acid is a somatically reconstituted nucleic acid or a clonotype that correlates with a precancerous condition or a cancerous condition such as a lymphoproliferative or myeloproliferative disorder. This can be used to monitor the disease state or condition. Such nucleic acids, and particularly clonotypes, are useful for monitoring minimal residual disease of cancer after treatment, and the results of such monitoring are important factors in determining continuation, interruption, or modification of treatment. It is. For many malignant lymphomas and bone marrow tumors, a diagnostic tissue sample, such as a peripheral blood sample or bone marrow sample, is obtained prior to the process of generating a clonotype profile ("diagnostic clonotype profile"). For lymphoproliferative or myeloproliferative disorders, it is usually not known before taking a diagnostic sample whether the immunoreceptor chain (s) are associated with a lymphocyte or bone marrow clone associated with the disease or condition. As a result, current practice separates amplification and sequencing for different reconstituted nucleic acids encoding different candidate immunoreceptor chains for the purpose of identifying clonotypes associated with the patient's disease or condition. It is necessary to carry out several times. In the chronotype profile generated by such laborious amplification and sequencing, one or more chronotypes related to the disease (ie, “related chronotypes”) are identified. Typically, the chronotype with the highest frequency in the chronotype profile is treated as the corresponding chronotype. In one aspect of the invention, by providing large-scale multiplex amplification in a single reaction of reconstituted nucleic acid portions encoding a plurality of different immunoreceptor chains, the relevant clonotypes can be identified. Significantly reduces the number of amplification and sequencing runs that were required. In some embodiments, “plurality” means that different immunoreceptor chains are within the range of 2-4; as well as in other embodiments, different from “plurality”. It means that the immunoreceptor chain is within the range of 2 to 3 types. More specifically, in some embodiments, among the BCR chains, recoding encoding the following: IgH containing at least part of the VDJ region, IgH containing at least part of the DJ region, and IgK. The constructed nucleic acids are amplified in a single multiplex reaction; and in other embodiments, among the TCR chains, reconstituted nucleic acids encoding the following: TCRβ, TCRδ and TCRγ are Amplify in a multiplex reaction.

  After treatment, and preferably after a complete remission of the cancer has been achieved, the presence, absence, or frequency of such associated clonotypes or nucleic acids is periodically evaluated in the post-treatment clonotype profile or nucleic acid profile. Based on the presence or increased frequency of the relevant nucleic acid or clonotype (or related clonotype), maintenance of remission or reappearance or recurrence of the tumor is diagnosed. That is, after treatment, minimal residual disease of the cancer is assessed based on the presence, absence, or frequency of the associated clonotype or characteristic nucleic acid. As noted above, when the related chronotype is common or equivalent to the reconstituted receptor fragment, there will be a lack of sufficient diversity (and thus non-cancerous cells) The presence of such clonotypes in the post-treatment clonotype profile can result in false positives indicating recurrence.

  The methods of the present invention are applicable to the monitoring of any proliferative disease in which reconstituted nucleic acids encoding immune receptors or portions thereof can be used as markers for cells involved in the disease. In one aspect, the methods of the invention are applicable to lymphoproliferative or myeloproliferative diseases. In another aspect, the methods of the invention are applicable to lymphomas and leukemias. In another aspect, the method of the invention comprises follicular lymphoma, chronic lymphocytic leukemia (CLL), acute lymphocytic leukemia (ALL), chronic myeloid leukemia (CML), acute myeloid leukemia (AML), Hodgkin and non- Hodgkin lymphoma, multiple myeloma (MM), benign monoclonal globulinemia (MGUS), mantle cell lymphoma (MCL), diffuse large B cell lymphoma (DLBCL), myelodysplastic syndrome (MDS), Alternatively, it can be applied to MRD monitoring in T cell lymphoma and the like. In certain embodiments, the method of the invention is good for monitoring ALL, MM, or DLBCL MRD.

In some embodiments, a diagnostic test may be performed on a patient sample such as blood or bone marrow and may include clonotypes produced by diseased clones (ie, related clonotypes) of the plurality of immunoreceptor chains. Identify things. Once the relevant clonotype of the immunoreceptor chain has been identified, subsequent monitoring assays can be made specific for the particular immunoreceptor chain. For example, in some embodiments, the diagnostic test generates a chronotype profile based on the sequence of multiple BCR chains such as IgH (VDJ), IgH (DJ), and IgK in the same reaction. If the relevant clonotype is an IgH (VDJ) chain, subsequent monitoring assays may generate only a clonotype profile of IgH (VDJ). In some embodiments, the depth of sequencing in the diagnostic sample can be different from that of the monitoring sample. “Sequencing depth” means the total number of sequence reads analyzed to construct a chronotype profile. For cancers such as leukemia or lymphoma, a diagnostic test is performed on a patient sample prior to treatment, and the frequency or level of the associated clonotype in the sample is typically high and easily identified. For example, any chronotype having a frequency that exceeds a predetermined level can be defined as an associated chronotype. Such a predetermined level may vary under other patient indicators; however, in many cases, the predetermined level may be in the range of 2-5%; or some embodiments Then, it may be 5%. Thus, in some embodiments, the depth at which sequencing is performed should be necessary to reliably detect clonotypes present at a frequency of 1 or 2% or more. In some embodiments, the depth of sequencing of the diagnostic sample is one that results in at least 10,000 sequence reads; or in other embodiments, it results in at least 100,000 sequence reads; In still other embodiments, the sequencing depth of the diagnostic sample is one that results in at least 10 ⁶ sequence reads. In some embodiments, the monitoring sample sequencing depth results in at least 100,000 sequence reads; in other embodiments, the monitoring sample sequencing depth is at least 10 ^6. This leads to a sequence read of

  In some embodiments, a lymphoproliferative disorder, such as a patient's leukemia or lymphoma, can be monitored by generating a clonotype profile from a sample (or tissue sample) obtained continuously from the patient. Such a chronotype profile can be generated as described above. In some embodiments, such monitoring comprises the following steps: (a) obtaining a sample comprising T cells and / or B cells and / or cell-free DNA from an individual; (b) T of the sample Adding a sequence tag to a reconstructed nucleic acid molecule of a cellular receptor gene or immunoglobulin gene to produce a tag-nucleic acid complex (wherein at least one reconstituted nucleic acid or a copy thereof is a different sequence) A tag is added); (c) amplifying the tag-nucleic acid complex; (d) sequencing the tag-nucleic acid complex sample, each having an error rate and the reconstructed nucleic acid Generating a sequence read comprising the sequence; (e) aligning sequence reads having similar tag sequences to form a group of sequence reads having the same sequence tag; (f Combining group sequence reads to determine clonotypes (where a group of sequence reads is bound to a different reconstituted nucleic acid sequence whenever the likelihood of the sequence read is clearly at least 95%) (G) determining the clonotype profile of the sample by determining the clonotype level; and (h) determining the level of the relevant clonotype in the clonotype profile. In some embodiments, steps (a)-(h) may be repeated in the process of monitoring the patient and diagnosing whether the associated clonotype level can be evidence of disease remission. In some embodiments, the adding and amplifying step comprises the following steps: (a) Reconstitution from immune cells expressing the first set of primers and immunoreceptors under primer extension conditions in the reaction mixture. Combining the prepared nucleic acid sample and / or cell-free DNA sample [wherein each primer of the first set has a receptor-specific portion, wherein the receptor-specific portion is Annealing at different positions and extending to generate a first extension product, each primer of the first set has a 5′-non-complementary end containing a first primer binding site (B) removing the first set of non-extended primers from the reaction mixture; and (c) adding a second set of primers to the reaction mixture under primer extension conditions. [Wherein each primer of the second set has a receptor-specific portion, wherein the receptor-specific portion anneals to the first extension product at a predetermined position, and Having a 5'-non-complementary end containing a primer binding site, wherein the first set of primers and / or the second set of primers comprises a receptor-specific portion and a first or second primer binding site, respectively; Each primer of the second set is extended to produce a second extension product, wherein each second extension product is bound to the first primer binding Site, a second primer binding site, at least one sequence tag, and a reconstituted nucleic acid encoding a portion of the immune cell receptor chain]. In some embodiments, the step of combining the reconstituted nucleic acid is whenever the likelihood of such sequence reads is clearly at least 99 percent; and in other embodiments, the likelihood is obvious This includes binding sequence reads of different reconstituted nucleic acids whenever at least 99.9 percent.

  The method of the present invention can also be applied to monitoring minimal residual disease in a patient's cancer, such as non-lymphoid or non-myeloid cancer, and the method can be applied, for example, in a selected set of oncogenes in a pattern of mutations. Including identifying. Such a mutation pattern, ie, the presence, absence, and / or level of a gene containing such a mutation may indicate the likelihood of disease recurrence. In some embodiments, target polynucleotides for such monitoring include exons, exon portions, selected introns and / or gene expression control regions, eg, promoters of multiple genes (herein “ Referred to as “oncogene molecule”). Oncogene molecules can be obtained by conventional methods such as exon capture methods [eg, TruSeq ™ exome enrichment kit (Illumina, San Diego, Calif.); Frampton et al, Nature Biotechnology, 31 (11): 1023-1031 (2013)]. Can be isolated from tissue samples. After obtaining such an oncogene molecule, a sequence tag is added to generate a tag-nucleic acid complex, and the tag-nucleic acid complex is amplified and sequenced according to the present invention.

  Recent cancer genome sequencing studies show that the mutation pattern is highly heterogeneous among different cancers, different patients suffering from the same cancer, cells of the same tumor, and cells of different metastases of the same patient However, in the same patient, because heterogeneous cancer cells typically evolve from a common ancestral cell, they share mutations and the evolutionary relationship between cancer cells is Can also be identified by continuous measurement over time [e.g., Vogelstein et al, Science, 339: 1546-1558 (2013); and Ding et al, Nature, 481 (7382): 506-510 ( Thus, the mutation pattern associated with cancer, as measured in a diagnostic sample, is the recurrence of the same cancer or the cancer To provide a Tedate for the detection of clonally advanced cancer.

  Oncogene molecules can be selected from a variety of genes, including but not limited to those listed in Table I.

  In some embodiments, the method of monitoring minimal residual disease of cancer described above comprises the steps of: (a) obtaining a tissue sample from an individual; (b) sequencing to a plurality of each oncogene molecule in the sample. Adding a tag to generate a tag-nucleic acid complex (wherein at least one nucleic acid or a copy thereof has a different sequence tag added and the oncogene molecule is characteristic of an individual's cancer (C) amplifying the tag-nucleic acid complex; (d) sequencing the tag-nucleic acid complex sample to form a sequence read having an error rate and including the tag sequence and the oncogene sequence. (E) aligning sequence reads having the same tag sequence to form a group of sequence reads having the same sequence tag; (f) combining the oncogene sequences of the group to oncogene Determining the sequence of the children, where the group of sequence reads is bound to a different oncogene molecule whenever the likelihood of the oncogene sequence group is clearly at least 95%; and (g) cancer Detecting in the gene molecule profile the presence, absence, and / or level of an oncogene molecule characteristic of an individual's cancer. In some embodiments, the step of joining the oncogene sequences is whenever such sequence reads differ by a likelihood of at least 99 percent; as well as in other embodiments, the likelihood varies by at least 99.9 percent Sometimes it involves joining sequence reads of different oncogene molecules.

Use of Sequence Tags to Detect Carryover Contamination Carryover contamination is a significant problem with techniques involving nucleic acid amplification [see, for example, Borst et al, Eur. J. et al. Clin. Microbiol. Infect. Dis. , 23 (4): 289-299 (2004); Aslanzadeh, Ann. Clin. Lab. Sci. , 34 (4): 389-396 (2004) etc.]. Such contamination occurs in sample assays when a small amount of nucleic acid that is not that of the sample is unintentionally amplified and affects the measurement results. In the worst case, carry-over contamination in a patient-derived medical sample may result in a false positive interpretation in the assay results. Nucleic acids that are not of a sample may be from a source unrelated to a particular patient; for example, may be from a sample from another patient. Alternatively, the non-sample nucleic acid may be from a patient-related source; for example, from a different sample from the same patient that was previously handled in the same laboratory Or from an assay reaction on another sample of the same patient that was previously processed in the same laboratory.

  Carryover contamination can be used in clinical settings when assessing highly complex populations of related nucleic acids, such as reconstituted nucleic acid groups encoding immune molecules such as T cell receptors or immunoglobulins. Especially problematic. The problem is that the sequence read or clonotype is an element of the actual diversity associated with the intended sample, or that they are a different nucleic acid source than the sample (in another patient sample, or in the same laboratory, This is because it is difficult to determine if it originates from a previous sample of the same patient that was processed in the same type of assay. In one embodiment of the present invention, not only the sequence tag is used to determine the clonotype based on the sequence read, but the sequence tag is derived from the current sample or from another sample. Such carry-over contamination can be detected by evaluating whether it is. Such detection maintains a record of the sequence tag determined from each patient sample, and then compares the latest measurement sequence tag to that previously measured whenever a subsequent measurement is made. Is achieved. Because of the large number of tags derived from each measurement and the ease of searching and comparing electronic records using conventional algorithms, such sequence tag records associated with clonotypes are recorded as electronic records. Conveniently maintained in mass storage. If the sequence tag group used for measurement is sufficiently large, it is very likely that carryover contamination has occurred when a match is found. A similar example for the size ratio of the sequence tag group to the clonotype group for labeling by sampling as described above is applicable to the detection of carryover contamination. In one embodiment, such ratio is 100: 1 or greater.

  Various search methods or algorithms can be used to perform the step of comparing the chronotype to be evaluated with the chronotype in the database. Many conventional sequence alignment methods and search algorithms are publicly available and are described in the following references incorporated by reference: Mount, Bioinformatics Sequence and Genome Analysis, Second Edition (Cold Spring). Harbor Press, 2004); Batzoglou, Briefings in Bioinformatics, 6: 6-22 (2005); Altschul et al, J. Biol. Mol. Biol. , 215 (3): 403-410 (1990); Needleman and Wunsch, J. Am. Mol. Biol. 48: 443-453 (1970); and Smith and Waterman, Advances in Applied Materials, 2: 482-489 (1981).

  In some embodiments, the above method for detecting and measuring contamination of a sample by materials from different samples, eg, contamination such as carryover contamination, comprises the following steps: (a) from a personal tissue sample (B) attaching a sequence tag to the oncogene molecule or reconstituted nucleic acid to produce a tag-nucleic acid complex (at least one nucleic acid or a copy thereof has a different sequence tag attached; The oncogene molecule is characteristic of an individual's cancer); (c) amplifying the tag-nucleic acid complex; (d) sequencing the tag-nucleic acid complex sample, each with an error rate. Providing a sequence read each comprising a tag sequence and an oncogene sequence or a reconstructed nucleic acid sequence; (e) the tag sequence is determined separately from other tissue samples Comparing with a tag sequence; and (f) evaluating the presence, absence, and / or degree of contamination thereby of one or more tag sequence identities with a tag sequence separately determined from other tissue samples. , May be included. Once the tag sequences have been determined in the assay, those tag sequences can also be compared to tag sequences in a database of tag sequences recorded by the assay for other patients. Such a comparison step can be performed at the time of the assay, or such a step can be performed retrospectively after the assay, for example. In one embodiment, the sequence tag is added to the reconstituted nucleic acid in a tissue sample (eg, blood or bone marrow) from an individual suffering from a lymphoproliferative disease, such as a cancer of lymphoid tissue. In another embodiment, sequence tags are added to oncogene molecules such as those described above.

  In a further embodiment where the reconstituted nucleic acid is monitored for cross-contamination of the tissue sample, the steps of adding and amplifying are as follows: (a) in the reaction mixture under primer extension conditions and with the first set of primers Combining a T cell-derived reconstituted nucleic acid sample and / or a cell-free DNA sample [wherein each primer of the first set has a receptor-specific portion, where the receptor specific The specific portion is annealed to a different reconstituted nucleic acid at a predetermined location and extended to produce a first extension product, each primer of the first set containing a first primer binding site. Having non-complementary ends]; (b) removing the first set of non-extended primers from the reaction mixture; (c) reaction mixing under primer extension conditions [Wherein each primer of the second set has a receptor-specific portion, wherein the receptor-specific portion has a first position at a predetermined position. Annealing the extension product and having a 5′-non-complementary end containing a second primer binding site, the first set of primers and / or the second set of primers are each a receptor specific portion and A sequence tag disposed between the first or second primer binding sites, and each primer of the second set extends to produce a second extension product, wherein each second The extension product comprises a first primer binding site, a second primer binding site, at least one sequence tag, and a reconstituted nucleic acid encoding a portion of an immunoreceptor chain], (d) a reaction mixture In Performing a polymerase chain reaction to produce an amplicon [the polymerase chain reaction comprises a forward primer specific for the first primer binding site and a reverse direction specific for the second primer binding site. Using a primer].

Kits The present invention includes various kits for performing the methods of the present invention. In some embodiments, the kit comprises (a) a set of forward primers and a set of reverse primers for amplifying a reconstituted nucleic acid encoding a plurality of immunoreceptor chains in multiplex PCR. [Where each forward primer and / or reverse primer has a target specific portion, a sequence tag, and a common primer binding site], (b) of the set of primers incorporated at least after the first extension A primer removal component for removing unprimed primer (ie, non-extended primer). In some embodiments, the kit further comprises a common primer specific for the common primer binding site. In some embodiments, the kit further comprises instructions for using the kit component in the methods of the invention. In some embodiments, the kit further comprises forward and reverse specific primers to amplify the reconstituted nucleic acid encoding IgH (VDJ), IgH (DJ), and IgK. . In some embodiments, the kit further comprises forward and reverse specific primers to amplify the reconstituted nucleic acid encoding TCRβ, TCRδ and TCRγ. In some embodiments, the kit further comprises an internal standard comprising a plurality of nucleic acids having a length and composition that reproduces the target reconstituted nucleic acid, wherein the internal standard is provided at a known concentration. In some embodiments, the kit includes a single stranded exonuclease such as E. coli exonuclease I as a primer removal component. In some embodiments, the kit includes a spin column that can sort double-stranded DNA based on size as a primer removal component.

  Although the present invention has been described with reference to specific examples for certain embodiments, those skilled in the art will recognize that many modifications can be made without departing from the spirit and scope of the invention. Let's go. The present invention is applicable to various sensor implementations and other subjects in addition to the above disclosure.

Definitions Unless otherwise stated, the terms and symbols relating to nucleic acid chemistry, biochemistry, genetics, and molecular biology used herein are standard papers and documents in the art, eg, Kornberg and Baker. , DNA Replication, Second Edition (W. H. Freeman, New York, 1992); New York, 1999); Abbas et al, Cellular and Molecular Imm. nology, is in accordance with the 6th edition (Saunders, 2007).

  “Alignment” is based on sequence distance measurement, which compares a test sequence, such as a sequence read, with one or more reference sequences to determine which reference sequence or which position of the reference sequence is closest. Means. An example of a method for aligning nucleotide sequences is the Smith Waterman algorithm. Examples of the distance measuring method include a Hamming distance or a Levenshtein distance. The distance measurement method can include elements related to the nucleotide quality score of the sequences to be compared.

  “Amplicon” means the product of a polynucleotide amplification reaction; ie, a clonal population of polynucleotides, which may be single-stranded or double-stranded nucleic acid, wherein the polynucleotide replicates from one or more starting sequences. Is done. The one or more starting sequences can be one or more copies of the same sequence, or they can be a mixture of different sequences. Amplicons can be generated by a variety of amplification reactions, the reaction product containing one or more copies of the starting or target nucleic acid. In one aspect, an amplification reaction that produces an amplicon is “performed on a template”. In this reaction, the base pairing of the reaction product, either nucleotide or oligonucleotide, is complementary to the template polynucleotide required for production of the reaction product. In one aspect, the reaction performed based on the template is primer extension with a nucleic acid polymerase or oligonucleotide ligation with a nucleic acid ligase. Such reactions include, but are not limited to, polymerase chain reaction (PCR), linear polymerase reaction, nucleic acid sequence-based amplification (NASBA), and rolling circle amplification. The following references which are incorporated herein by reference: Mullis et al, US Pat. Nos. 4,683,195; 4,965,188; 4,683,202; 4,800,159 (PCR); Gelfand et al, US Pat. No. 5,210,015 (real-time PCR with “taqman” probe); Wittwer et al, US Pat. No. 6,174,670; Kacian et al. US Patent No. 5,399,491 ("NASBA"); Lizardi, US Patent 5,854,033 Patent; Aono et al, is disclosed in, Japanese Patent Laid-Open No. 4-262799 (rolling circle amplification). In one aspect, the amplicons of the invention are generated by PCR. When a detection chemical reaction capable of measuring a reaction product as the amplification reaction proceeds is available, for example, the following “real-time PCR” or Leone et al, Nucleic Acids Research, 26: 2150-2155 (1998). ) And similar references, such as “real-time NASBA” can also be “real-time” amplification. As used herein, the term “amplification” means performing an amplification reaction. “Reaction mixture” means a solution containing all of the reactants necessary to carry out the reaction, such as, but not limited to, a pH selected to the extent selected during the reaction. Buffers, salts, cofactors and scavengers to maintain can be mentioned.

  As used herein, “clonality” means a measure that, in the repertoire chronotype, slopes the distribution of chronotype abundance to a single or a few chronotypes. Broadly speaking, clonality is a measure opposite to chronotype diversity. From ecology, many measures and statistics are available to explain the species richness relationship. For example, according to the idea of Pierou, “An Introduction to Mathematical Ecology”, Chapters 17 and 18 (Wiley-Interscience, 1969), using measures and statistics to explain the relationship Can evaluate clonality. In one aspect, the measure of clonality used in the present invention is a function of the chronotype profile (ie, the number of chronotypes to be detected and their abundance), and after evaluation of the chronotype profile, the chronotype You can also calculate clonality from a profile to get a single number. One clonality measure is simply the Simpson measure, which determines the probability that two randomly selected chronotypes are the same. Other clonality measures include the information-based measures disclosed in Pielou (supra) and the McIntosh diversity index.

  “Chronotype” means a reconstituted nucleic acid of lymphocytes that encodes an immunoreceptor or portion thereof. More specifically, a clonotype is usually extracted from a T cell or B cell and optionally encodes a T cell receptor (TCR) or B cell receptor (BCR) or a portion thereof. It refers to a reconstituted nucleic acid that can be derived from a cell-free resource. In various embodiments, the clonotype is an IgH VDJ rearrangement, IgH DJ rearrangement, IgK VJ rearrangement, IgL VJ rearrangement, TCRβ VDJ rearrangement, TCRβ DJ rearrangement It may encode all or part of a sequence, a TCRα VJ rearrangement sequence, a TCRγ VJ rearrangement sequence, a TCRδ VDJ rearrangement sequence, a TCRδ VD rearrangement sequence, or a Kde-V rearrangement sequence. The clonotype may also encode a translocation cleavage region that includes an immunoreceptor gene, such as Bcl-IgH or Bel I-IgH. In one aspect, clonotypes have sequences that are long enough to reproduce or reflect the diversity of the immune molecules from which they are derived; as a result, the length of clonotypes can vary greatly. In some embodiments, the clonotype may have a length in the range of 25-400 nucleotides; and in other embodiments, the clonotype has a length in the range of 25-200 nucleotides.

“Chronotype profile” means enumerating various chronotypes and relative abundances from a lymphocyte population, eg, relative abundance can be expressed as a frequency in a given population (Ie a number between 0 and 1). Typically, the lymphocyte population is obtained from a tissue sample. The term “chronotype profile” refers to the following: Arstila et al, Science, 286: 958-961 (1999); Yassai et al, Immunogenetics, 61: 493-502 (2009); and Kedzierska et al, Mol. Immunol. , 45 (3): 607-618 (2008) and related to the immunity concept for the immunity “repertoire” as described in references, but is more general. The term “chronotype profile” includes a variety of lists and abundance of reconstituted nucleic acids encoding immunoreceptors, such profiles can be selected for subsets of lymphocytes (eg, tissue infiltrating lymphocytes, or immune lymphocytes). Such as a subset of phenotypes) or may encode immunoreceptor moieties that have reduced diversity as compared to the full immune receptor. In some embodiments, a chronotype profile may include at least 10 ³ different chronotypes; in other embodiments, a chronotype profile may include at least 10 ⁴ different chronotypes; In form, the chronotype profile may include at least 10 ⁵ different chronotypes; in other embodiments, the chronotype profile may include at least 10 ⁶ different chronotypes. In such embodiments, such chronotype profiles can further include abundance or relative frequency for each different chronotype. In one aspect, the clonotype profile has different (abundant) recombinant nucleotides encoding T cell receptor (TCR) or B cell receptor (BCR) or fragments thereof, respectively, in an individual lymphocyte population. A set of sequences, wherein the nucleotide sequences of the set relate to substantially all of the lymphocytes of the population and have a one-to-one correspondence with different lymphocytes or their clonal subpopulations. In one aspect, chronotypic diversity (ie, different nucleic acid sequences in a set) to allow substantially all T cells or B cells of an individual or clones thereof to transport the unique nucleic acid sequence of such a repertoire. Select a nucleic acid segment that defines the clonotype. That is, preferably each different clone of the sample has a different clonotype. In other aspects of the invention, the lymphocyte population corresponding to repertoire may be circulating B cells, or circulating T cells, or the aforementioned population, such as but not limited to CD4 + T cells or CD8 + T cells. Or any other subpopulation defined by cell surface markers. Such subpopulations can also be obtained by taking samples from specific tissues, such as bone marrow or lymph nodes, or based on one or more cell surface markers, sizes, or forms, etc. It can also be obtained by sorting or enriching cells from (eg peripheral blood). In yet another aspect, the lymphocyte population corresponding to repertoire may be derived from a diseased tissue such as a tumor tissue or an infected tissue. In one embodiment, a clonotype profile comprising a human TCR β chain or a fragment thereof has different nucleotide sequences ranging from 0.1 × 10 ^{6 to} 1.8 × 10 ⁶ , or 0.5 × 10 ^{6 to} 1.5. × 10 ⁶ range, or a number in the range of 0.8 × ¹⁰ 6 to 1.2 × 10 ^6. In another embodiment, a clonotype profile comprising a human Ig heavy chain or a fragment thereof has a different nucleotide sequence ranging from 0.1 × 10 ^{6 to} 1.8 × 10 ⁶ , or 0.5 × 10 ⁶ to 1. 5x10 ⁶ range, or a number in the range of ^{0.8 × 10 6 ~1.2 × 10 6} . In certain embodiments, a clonotype profile of the invention comprises a set of nucleotide sequences that encode substantially all of the segments of the V (D) J region of the Ig heavy chain. As used herein, in one aspect, “substantially all” means all of the segments that have a relative abundance greater than or equal to 0.001%; or, as used herein, In the embodiment, “substantially all” means all of the segments whose relative abundance is 0.0001% or more. In another specific embodiment, the clonotype profile of the invention comprises a set of nucleotide sequences encoding substantially all segments of the V (D) J region of the TCR beta chain. In another embodiment, the clonotype profile of the present invention comprises a set of nucleotide sequences that range in length from 25 to 200 bases in nucleotides and include segments of the V, D, and J regions of the TCR beta chain. In another embodiment, a clonotype profile of the present invention comprises a set of nucleotide sequences that range in length from 25 to 200 bases in nucleotides and include segments of the V, D, and J regions of the Ig heavy chain. In another embodiment, the clonotype profile of the invention comprises a number of different nucleotide sequences, this number being substantially equal to the number of lymphocytes expressing different Ig heavy chains. In another embodiment, the clonotype profile of the present invention comprises a number of different nucleotide sequences, this number being substantially equal to the number of lymphocytes expressing different TCRβ chains. In yet another embodiment, “substantially equal” encodes an IgH or TCRβ or portion thereof whose chronotype profile is transported or expressed with a frequency of 0.001% or more by all lymphocyte populations of the individual. It means that it has a 99% probability. In yet another embodiment, “substantially equal” encodes IgH or TCRβ or a portion thereof whose repertoire nucleotide sequence is transported or expressed by all lymphocyte cells present at a frequency of 0.0001% or greater. It means that it has a 99% probability. In some embodiments, the clonotype profile is derived from a sample composed of 10 ⁵ to 10 ⁷ lymphocytes. Such lymphocyte counts can be obtained from 1 to 10 mL of peripheral blood samples.

  A “complementarity-determining region” (CDR) is a region of an immunoglobulin (ie, antibody) or T cell receptor that determines its specificity by complementing the conformation of the antigen. And a region in contact with a specific antigen. T cell receptors and immunoglobulins each have three CDRs: CDR1 and CDR2 are found in the variable (V) domain, CDR3 has some V domain and all diversity (D) domains (heavy chain only ) And a linking domain (J) and some constant (C) domain.

“Chronotype database” means a chronotype collection that has been formatted and prepared for ease of searching, comparison, and retrieval and speed. In some embodiments, the clonotype database includes a collection of clonotypes that encode the same region or segment of the immunoreceptor. In some embodiments, the clonotype database includes clonotype profiles of clonotypes from multiple individuals. In some embodiments, the chronograph type database includes anticlonotype profile of at least 10 ⁴ clonotypes from individuals of at least 10 people. In some embodiments, the chronotype database includes at least 10 ⁶ chronotypes, or at least 10 ⁸ chronotypes, or at least 10 ⁹ chronotypes, or at least 10 ¹⁰ chronotypes. A public database including a chronotype such as an IMGT database (www.imgt.org) may be a chronotype database, and is described in, for example, Nucleic Acids Research, 31: 307-310 (2003). The chronotype database may be in the FASTA format and may be BLAST algorithms, such as Altschul et al, J. MoI. Mol. Biol. , 215 (3): 403-410 (1990), etc., can be used to search or compare entries in the chronotype database.

  "Binding" is to judge two clonotype candidates with different sequences as such differences are due to empirical and experimental errors and not really due to biological differences. Means that the same processing is performed. In one aspect, the sequence of high frequency chronotype candidates is compared to that of low frequency chronotype candidates, and if the predetermined criteria are met, the number of low frequency chronotype candidates is increased to the high frequency chronotype candidate. In addition to the number of type candidates, this low frequency chronotype candidate is ignored thereafter. That is, the read count associated with the low frequency chronotype candidate is added to the high frequency chronotype candidate read count, and the high frequency chronotype candidate and the low frequency chronotype candidate are treated as the same thing; It is determined that the observed difference is due to an error (eg, sequencing error or amplification error). In some embodiments, the predetermined criterion is a likelihood function that depends on factors such as the relative frequency of the chronotype candidates to be compared, the number of different positions of the candidates, and the quality score of the positions.

  A “complementarity-determining region” (CDR) is a region of an immunoglobulin (ie, antibody) or T cell receptor that determines its specificity by complementing the conformation of the antigen. And a region in contact with a specific antigen. T cell receptors and immunoglobulins each have three CDRs: CDR1 and CDR2 are found in the variable (V) domain, CDR3 has some V domain and all diversity (D) domains (heavy chain only ) And a linking domain (J) and some constant (C) domain.

  As used herein, “contamination” means the presence of nucleic acid from another individual in a tissue sample of one individual. In one aspect, “contamination” means that there is non-patient-derived nucleic acid that can affect the interpretation of the patient's clonotype profile.

  “Identification by gene” refers to a unique correspondence between an individual and a set of values (or statuses) of genomic markers derived from one or more loci of the individual.

  “Gene marker” means a polymorphic segment of DNA at a locus that can be used to identify an individual. Genetic markers can also be identified based on sequence or based on flanking or flanking sequences. Typically, a genetic marker can have multiple sequences or values in each individual of the group. Examples of genomic markers include, but are not limited to, tandem repeats (STR) and single nucleotide polymorphisms (SNPs). The polymorphic fragment of DNA may be genomic DNA or reverse transcribed RNA. In one embodiment, the polymorphic fragment is genomic DNA. In one embodiment, genetic markers used in the present invention are identified by amplification and sequencing utilizing conventional methods. In another embodiment, the genetic marker is amplified and sequenced with the immune molecule during the process for generation of the clonotype profile.

“Internal standard” means a nucleic acid sequence that is processed in the same reaction as one or more target polynucleotides for the purpose of quantifying the absolute or relative amount of a target polynucleotide in a sample. In one aspect, the reaction is an amplification reaction such as PCR. The internal standard may be intrinsic or in addition. That is, the internal standard can be naturally occurring in the sample or it can be added to the sample prior to the reaction. In one aspect, a calibration is obtained by adding one or more internal standard sequences to be added to the reaction mixture at a predetermined concentration, and comparing the amplified sequences to this calibration allows the corresponding target poly The amount of nucleotides can be determined. The number, arrangement, length, and other features of the internal standard to add are routine choices for those skilled in the art at design time. An endogenous internal standard, also referred to herein as a “reference sequence”, is a sequence that is native to the sample, with minimal control, and at a constant and cell cycle-dependent level of transcription. The gene representing the product corresponds to [eg, Selvey et al, Mol. Cell Probes, 15: 307-311 (2001)]. Examples of endogenous internal standards include, but are not limited to, sequences derived from the following genes: GAPDH, β ₂ -microglobulin, 18S ribosomal RNA, and β-actin.

  “Kit” refers to any delivery system for delivering materials or reagents to perform a method of the invention. In the context of the method of the present invention, such a delivery system is a reaction reagent from one location to another (eg, primer, enzyme, internal standard, etc. in a suitable container) and / or a support material (eg, buffer). Agents, instructions for performing the assay, etc.) including systems that allow storage, transport or delivery. For example, the kit includes one or more containers (eg, boxes) that contain relevant reaction reagents and / or indicator materials. Such contents can be provided in conjunction with or separately from the intended container. For example, the first container may contain an enzyme used for the assay, and the second container may contain a primer.

  “Minor residual disease” means cancer cells remaining after treatment. The term is frequently used in connection with lymphoma and leukemia.

By “lymphoproliferative disorder or myeloproliferative disorder” is meant any aberrant proliferative disorder, for monitoring such a disorder one that encodes one or more reconstituted immune receptors. The above nucleotide sequence can be used as a marker. A “lymphoid or bone marrow tumor” refers to an abnormal growth of lymphocytes or bone marrow cells and can be malignant or non-malignant. A cancer of the lymphoid tissue is a malignant lymph tumor. Bone marrow cancer is a malignant myeloma. Lymphoid and bone marrow tumors are the result of or associated with lymphoproliferative or myeloproliferative diseases, including but not limited to follicular lymphoma, chronic lymphocytic leukemia (CLL), acute lymph Leukemia (ALL), chronic myeloid leukemia (CML), acute myeloid leukemia (AML), Hodgkin and non-Hodgkin lymphoma, multiple myeloma (MM), benign monoclonal gamma globulinemia (MGUS), mantle cells Lymphoma (MCL), diffuse large B-cell lymphoma (DLBCL), myelodysplastic syndrome (MDS), or T-cell lymphoma, eg, Jaffe et al, Blood, 112: 4384-4399 (2008); al, WHO Classification of Tumours of Haema ^{opoietic and Lymphoid Tissues (e.4 th)} (IARC Press, 2008) , and the like. As used herein, “B cell cancer” refers to a lymphoid or bone marrow tumor involving B cells or cells that differentiate from B cells, such as plasma cells. Similarly, “T cell carcinoma” refers to a lymphoid or bone marrow tumor involving T cells or cells that differentiate from T cells.

  Terms such as “% homology” or “% identity” used in reference to a comparison of a reference sequence and another sequence (“comparison sequence”) refer to those in optimal alignment between the two sequences. In the comparison sequence, the subunit part is the same as that of the reference sequence in the stated ratio. Subunits are nucleotides when comparing polynucleotides or amino acids when comparing polypeptides. As used herein, “optimal alignment” of the sequences to be compared is one that maximizes the match between subunits and minimizes the number of gaps used to construct the alignment. % Identity is determined by Needleman and Wunsch, J. et al. Mol. Biol. , 48: 443-453 (1970) ("GAP" program of Wisconsin Sequence Analysis Package, Genetics Computer Group, Madison, WI). . Other software packages for building alignment and calculating% identity in the industry, or other similar measurement methods, include program, based on the algorithm of Smith and Water, Advanced in Applied Mathematics, 2:48. ~ 489 (1981) (Wisconsin Sequence Analysis Package, Genetics Computer Group, Madison, Wis.) "BestFit" program. In other words, for example, to obtain a polynucleotide having a nucleotide sequence that is at least 95% identical to a reference nucleotide sequence, up to 5% nucleotides in the reference sequence are deleted or replaced by another nucleotide. Alternatively, up to 5% of the number of nucleotides in the total number of nucleotides in the reference sequence can be inserted into the reference sequence.

“Polymerase chain reaction” or “PCR” means a reaction that amplifies a specific DNA sequence in vitro by facilitating simultaneous primer extension of complementary DNA strands. In other words, PCR is a reaction for producing a plurality of copies or replicas of a target nucleic acid adjacent to a primer binding site. In such a reaction, the following steps are performed: Including annealing the primer to the primer binding site and (iii) extending the primer with a nucleic acid polymerase in the presence of a nucleoside triphosphate one or more times. As used herein, the terms “forward primer” and “upstream primer” are used interchangeably, and the terms “reverse primer” and “downstream primer” are used interchangeably. Similarly, as used herein, when the double-stranded target polynucleotide is shown in a 5 ′ → 3 ′ orientation with the sense strand left to right, the forward primer is on the left side of the antisense strand. Combine and extend to the right, reverse primer binds to the right side of the sense strand and extends to the left. Usually, in the reaction, different temperatures optimized for each step are periodically repeated in a thermal cycler. The specific temperature, the duration of each step, and the rate at which each step is taken are a number of factors well known to those skilled in the art, such as reference: McPherson et al, editors, PCR: A Practical Approach and. PCR2: A Practical Approach (IRL Press, Oxford, 1991 and 1995, respectively): For example, in a conventional PCR using Taq DNA polymerase, a double-stranded target nucleic acid is denatured at a temperature above 90 ° C., and a primer is annealed at a temperature in the range of 50-75 ° C. Extend the primer at a range of temperatures. The term “PCR” encompasses derivative systems of reactions such as, but not limited to, RT-PCR, real-time PCR, nested PCR, quantitative PCR, and multiplexed PCR. The reaction dose ranges from several hundred nanoliters, for example 200 nL to several hundred μL, for example 200 μL. “Reverse transcription PCR” or “RT-PCR” means PCR that proceeds by a reverse transcription reaction in which target RNA is converted to complementary single-stranded nucleic acid DNA and then amplified [eg, herein by reference. Tecott et al, US Pat. No. 5,168,038, which is incorporated herein by reference. “Real-time PCR” means PCR in which the amount of reaction product, that is, the amount of amplicon is monitored as the reaction proceeds. There are many forms of real-time PCR where the chemical mechanism of the detection method utilized to monitor the reaction product is the main difference [eg, Gelfand et al., Incorporated herein by reference. al., US Pat. No. 5,210,015 (“taqman”); Wittwer et al, US Pat. Nos. 6,174,670 and 6,569,627 (intercalating dyes); Tyagi et al. US Pat. No. 5,925,517 (molecular beacon)]. The detection chemistry of real-time PCR has been reviewed by Mackay et al, Nucleic Acids Research, 30: 1292-1305 (2002), which is incorporated herein by reference. “Nested PCR” means a two-stage PCR, in which the amplicon of the first PCR is a sample of the second PCR, and in the second PCR, at least one primer is the first amplicon. Use a new primer set that binds to the inner region. As used herein, in reference to a nested amplification reaction, “first primer” refers to the primer utilized to generate the first amplico, and “second primer” refers to the second Or one or more primers used to generate a nested amplicon. “Multiplexed PCR” refers to PCR in which multiple target sequences (or a single target sequence and one or more reference sequences) are used simultaneously in the same reaction mixture [see, eg, Bernard et al, Anal. Biochem. 273: 221-228 (1999) (two-color real-time PCR)]. Usually, a separate primer set is used for each sequence to be amplified. “Quantitative PCR” means a PCR designed to assess the abundance of one or more specific target sequences in a sample or specimen. Quantitative PCR includes both full and relative amounts of target sequence. Quantitative measurements are made utilizing one or more reference sequences or internal standards that can be assayed separately or together with the target sequence. The reference sequence may be endogenous to the sample or specimen, or may be added from others, and in the latter case, one or more competing templates may be included. Typical endogenous reference sequences include transcript segments such as the following genes: β-actin, GAPDH, β ₂ -microglobulin, and ribosomal RNA. Quantitative PCR techniques are incorporated herein by reference, the following references: Freeman et al, Biotechniques, 26: 112-126 (1999); Becker-Andre et al, Nucleic Acids Research, 17: 9437. -9447 (1989); Zimmerman et al, Biotechniques, 21: 268-279 (1996); Divacco et al, Gene, 122: 3013-3020 (1992); Becker-Andre et al, Nucleic Acids Research 37: 9446 (1989) and the like are well known to those skilled in the art.

A “primer” can function as a starting point for nucleic acid synthesis by forming a duplex with a polynucleotide template and can extend from the 3 ′ end along the template to form an extended duplex. Natural or synthetic oligonucleotide. Primer extension is usually performed with a nucleic acid polymerase such as DNA or RNA polymerase. The nucleotide sequence added to the extension process is determined based on the sequence of the template polynucleotide. Usually, the primer is extended by a DNA polymerase. A primer usually has a length in the range of 14 to 40 bases of nucleotides, or a length in the range of 18 to 36 bases of nucleotides. The primer is used in various nucleic acid amplification reaction methods (for example, linear amplification reaction using a single primer, or polymerase chain reaction using two or more primers). Pertain to certain applications, the guidelines for selecting the length and sequence of the primer, the following references are incorporated by ^{reference: Dieffenbach, editor, PCR Primer:} A Laboratory Manual, 2 nd Edition (Cold Spring Harbor Press, New York, As is proved by 2003).

  “Quality score” means a measure of the probability that the assignment of a base at a particular sequence position is appropriate. For those skilled in the art, various techniques for calculating a quality score (eg, base call, etc.) are well known to those skilled in the art for a particular condition, for example, bases that are called as a result of different sequencing chemistry, detection systems, and base call algorithms, etc. Are known. In general, the quality score value is monotonically related to the probability that the base call is correct. For example, if the quality score or Q is 10, it means that the probability that the base is correctly called is 90%, and if Q is 20, it means that the probability that the base is called correctly is 99%. This also applies sequentially to other cases as well. For some sequencing platforms, particularly those that use synthetic decoding, the average quality score is incomplete extension, possessing forward extension, template loss, polymerase loss, protection failure, and loss of Due to a phenomenon such as failure of protection, the quality score at the start point of the sequence read becomes higher than the quality score at the end point of the sequence read because it decreases as the sequence read becomes longer.

  “Sequence read” means a nucleotide sequence determined from a sequence or data stream generated by a sequencing method, which is determined by, for example, base call software related to the technique, eg, a private provider of a DNA sequencing platform It is made by a method using base call software. A sequence read usually includes a quality score for each nucleotide in the sequence. Typically, sequence reads are made by extending primers along a template nucleic acid, for example, with DNA polymerase or DNA ligase. Data is generated based on recorded signals such as optical signals, chemical signals (eg, pH changes), or electrical signals associated with such elongation. Such initial data is converted into an array read.

  A “sequence tag” (or “tag”) or “barcode” is an oligo that is added to a polynucleotide or template molecule and used to identify and / or track a polynucleotide or template in a reaction or series of reactions. Means nucleotides. Each sequence tag has a nucleotide sequence often referred to herein as a “tag sequence”. A sequence tag can be added to the 3 ′ or 5 ′ end of a polynucleotide or template, or can be inserted inside a polynucleotide or template and optionally referred to herein as a “tagged polynucleotide” or “tagged template”. Or a “tag-polynucleotide complex”, a “tag-molecule complex”, or the like can be formed. The size and composition of the sequence tag can vary widely; the following references, which are hereby incorporated by reference: Brenner, US Pat. No. 5,635,400; Brenner and Macevicz, US Pat. No. 7, 537, 897; Brenner et al, Proc. Natl. Acad. Sci. 97: 1665-1670 (2000); Church et al, European Patent Publication No. 0 303 459; Shoemaker et al, Nature Genetics, 14: 450-456 (1996); Wallace, US Pat. No. 5,981,179, etc. provides guidance for selecting a suitable sequence tag set for a specific embodiment. Specific tag length and / or composition choices include, but are not limited to, the sequencing method used to decode the tag; the distinguishable required to unambiguously identify the target polynucleotide set To ensure reliable identification, the number of tags depends on several factors such as the degree to which the required tags of the set differ, for example, freedom from misidentification due to cross-hybridization or sequencing errors. obtain. In some embodiments, the sequence tag is in the range of 6-100 bases of nucleotides, or in the range of nucleotides 10-100 bases, or in the range of nucleotides 12-50 bases, or in the range of nucleotides 12-25 bases, respectively. Has a length of within. In some embodiments, a sequence tag set is used in which each sequence tag in the set has a unique nucleotide sequence that differs by at least 4 bases from all other tags in the same set; in other embodiments, each set tag A sequence tag set is used in which the sequence tag differs from all other tags in the same set by at least 5 bases; in still other embodiments, each sequence tag in the set has at least all other tags and bases in the same set. Sequence tag sets that are 10% different are used; or, in other embodiments, sequence tag sets that are at least 25% different bases are used; or, in other embodiments, sequence tag sets that are at least 50% different bases. Is used.

Claims (31)

A method for generating a clonotype profile from a plurality of immune cell receptor chains comprising:
(A) combining a first set of primers with a reconstituted nucleic acid sample and / or cell-free DNA sample from B cells and / or T cells under primer extension conditions in a reaction mixture, Each primer of the first set has a receptor specific portion, a 5′-non-complementary end containing a first primer binding site, and the receptor specific portion and the first primer binding site. The receptor-specific portion is annealed to a different reconstituted nucleic acid at a first predetermined position and extended to produce a first extension product. Process and;
(B) removing a non-extended primer from the first mixture of primers from the reaction mixture;
(C) adding a second set of primers to the reaction mixture under primer extension conditions, each primer of the second set comprising a receptor-specific portion and a second primer binding site. The receptor-specific portion anneals to the first extension product at a second predetermined position, and each primer of the second set is extended. To generate a second extension product, each second extension product comprising a first primer binding site, a second primer binding site, a sequence tag, and a T cell receptor chain or a B cell receptor chain. Comprising a reconstituted nucleic acid encoding portion; and
(D) performing a polymerase chain reaction in the reaction mixture to generate an amplicon, the polymerase chain reaction comprising a forward primer specific for the first primer binding site; Using a reverse primer specific for the primer binding site of
(E) sequencing the amplicon to generate a clonotype profile of a plurality of T cell receptor chains and / or B cell receptor chains, from a plurality of immune cell receptor chains. How to generate a profile.   The plurality of T cell receptor chains comprise TCRβ, TCRδ and TCRγ, and the first set of primers and the second set of primers are in a reconstituted nucleic acid region encoding a VDJ region of TCRβ and TCRδ. The method of claim 1, comprising an adjacent primer and a primer adjacent to the VJ region of TCRγ. The sequencing step includes:
Generating sequence reads having respective error rates and comprising respective tag sequences and reconstructed nucleic acid sequences;
Aligning sequence reads having the same tag sequence to form a group of sequence reads having the same sequence tag, each group of sequence reads being a candidate clonotype;
Determining a different clonotype, wherein the relative frequency or read count in the group of sequence reads to be compared exceeds a predetermined threshold and the number of different nucleotides in the group of sequence reads to be compared is When each candidate chronotype is clearly at least 95% likely compared to the other candidate chronotypes when a defined threshold is exceeded , binding the sequence read to a different chronotype To decide,
The method of claim 1 comprising:   The method of claim 1, further comprising removing a non-extended second set of primers after generation of the second extension product.   The method of claim 1, wherein annealing and extension of the first set of the primers is repeated after melting of the first extension product.   The method of claim 1, wherein annealing and extension of the second set of the primers is repeated after melting of the second extension product.   5. The method of claim 4, wherein the removing step comprises adding an enzyme having exonuclease activity to the reaction mixture.   The plurality of B cell receptor chains include IgH and IgK, and the first set of primers and the second set of primers encode an IgH VDJ region, an IgH DJ region, and an IgK VJ region. The method of claim 1, comprising a primer adjacent to the reconstituted nucleic acid region. A method for generating a clonotype profile of a reconstituted nucleic acid encoding a plurality of immunoreceptor chains in a sample comprising:
(A) Tag-nucleic acid by adding a sequence tag to a reconstructed nucleic acid molecule of a T cell receptor gene or an immunoglobulin gene of a sample derived from an individual containing T cells and / or B cells and / or cell-free DNA Creating a complex, wherein at least one tag-nucleic acid complex or a copy thereof has a different sequence tag added thereto;
(B) amplifying the tag-nucleic acid complex;
(C) sequencing the tag-nucleic acid complex to provide a sequence read having a respective error rate and comprising each sequence tag and the reconstructed nucleic acid sequence;
(D) aligning sequence reads having the same sequence tag to form a group of sequence reads having the same sequence tag, each group of sequence reads being a candidate clonotype Process and;
(F) determining the different clonotypes in the sample, each relative frequency or read count exceeding a predetermined threshold and each candidate when the number of different nucleotides exceeds a predetermined threshold Generate a clonotype profile for the sample by determining that the sequence read binds to a different clonotype if that clonotype is clearly at least 95% likely compared to other candidate clonotypes And a process of
Only including,
The adding step includes
Combining a first set of primers and a sample under primer extension conditions in a reaction mixture, wherein each primer of the first set contains a receptor-specific portion, a first primer binding site. 5′-non-complementary ends, and a sequence tag disposed between the receptor-specific portion and the first primer binding site, the receptor-specific portion comprising a first predetermined Annealing to different reconstituted nucleic acids at a position and extending to produce a first extension product; combining;
Removing an unextended primer of the first set of primers from the reaction mixture;
Adding a second primer set to the reaction mixture under primer extension conditions, each primer of the second set comprising a receptor-specific portion and a second primer binding site 5. The receptor-specific portion has a non-complementary end and anneals to the first extension product at a second predetermined position, and each primer of the second set is extended A second extension product, each second extension product comprising a first primer binding site, a second primer binding site, a sequence tag, and a portion of a T cell receptor chain or B cell receptor chain. Including, including, adding reconstituted nucleic acid encoding,
A method of generating a clonotype profile of a reconstituted nucleic acid encoding a plurality of immunoreceptor chains in a sample. The method of claim 9 , further comprising removing the second set of non-extended primers from the reaction mixture. 10. The method of claim 9 , wherein the annealing and extension of the first set of the primers is repeated after melting of the first extension product. 10. The method of claim 9 , wherein the annealing and extension of the second set of primers is repeated after melting of the second extension product. 11. The method of claim 10 , wherein the removing step comprises adding an enzyme having exonuclease activity to the reaction mixture.   10. The adding and the amplifying steps generate a tag-nucleic acid complex, wherein a plurality of different sequence tags are added to at least one reconstituted nucleic acid and its respective copy. the method of. The reconstituted nucleic acid comprises a reconstituted nucleic acid encoding TCRβ, TCRδ and TCRγ chains, and the first set of primers and the second set of primers encode VCR regions of TCRβ and TCRδ. 15. The method of claim 14 , comprising a primer adjacent to the rearranged nucleic acid region and a primer adjacent to the VCR region of TCRγ. The reconstituted nucleic acid comprises a reconstituted nucleic acid encoding IgH and IgK, and the first set of primers and the second set of primers comprise an IgH VDJ region, an IgH DJ region, and 15. The method of claim 14 , comprising a primer adjacent to the rearranged nucleic acid region encoding the IgK VJ region.   The method of claim 1, wherein the sequence tag is a mosaic tag, and wherein the mosaic tag includes alternating constant regions and variable regions.   2. The method of claim 1, wherein the forward primer and / or reverse primer comprises a 5'-non-complementary end containing a sample tag.   The method of claim 1, wherein the first predetermined position is a V region. 20. The method of claim 19 , wherein the receptor specific portion of each primer of the first set anneals to a different non-overlapping portion of the V region.   The method according to claim 1, wherein the second predetermined position is a J region or a C region. 22. The method of claim 21 , wherein the receptor specific portion of each primer of the second set anneals to a different non-overlapping portion of the J region. 22. The method of claim 21 , wherein the receptor specific portion of each primer of the second set anneals to a single primer binding site in the J region. The adding step comprises performing a polymerase chain reaction in the reaction mixture to generate an amplicon, wherein the polymerase chain reaction comprises a forward primer specific for the first primer binding site; and 10. The method of claim 9 , wherein a reverse primer specific for the second primer binding site is used. 25. The method of claim 24 , wherein the forward primer and / or reverse primer comprises a 5'-non-complementary end containing a sample tag.   The method of claim 9, wherein the sequence tag is a mosaic tag, and wherein the mosaic tag includes alternating constant regions and variable regions. The method of claim 9 , wherein the first predetermined position is a V region. 28. The method of claim 27 , wherein the receptor specific portion of each primer of the first set anneals to a different non-overlapping portion of the V region. The method according to claim 9 , wherein the second predetermined position is a J region or a C region. 30. The method of claim 29 , wherein the receptor specific portion of each primer of the second set anneals to a different non-overlapping portion of the J region. 30. The method of claim 29 , wherein the receptor specific portion of each primer of the second set anneals to a single primer binding site in the J region.

Images